[Federal Register: January 24, 2008 (Volume 73, Number 16)]
[Rules and Regulations]
[Page 4311-4377]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr24ja08-32]
[[Page 4311]]
-----------------------------------------------------------------------
Part II
Environmental Protection Agency
-----------------------------------------------------------------------
40 CFR Parts 72 and 75
Revisions to the Continuous Emissions Monitoring Rule for the Acid Rain
Program, NOX Budget Trading Program, Clean Air Interstate
Rule, and the Clean Air Mercury Rule; Final Rule
[[Page 4312]]
-----------------------------------------------------------------------
ENVIRONMENTAL PROTECTION AGENCY
40 CFR Parts 72 and 75
[EPA-HQ-OAR-2005-0132; FRL-8511-1]
RIN 2060-AN16
Revisions to the Continuous Emissions Monitoring Rule for the
Acid Rain Program, NOX Budget Trading Program, Clean Air
Interstate Rule, and the Clean Air Mercury Rule
AGENCY: Environmental Protection Agency (EPA).
ACTION: Final rule.
-----------------------------------------------------------------------
SUMMARY: EPA is finalizing rule revisions that modify existing
requirements for sources affected by the federally administered
emission trading programs including the NOX Budget Trading
Program, the Acid Rain Program, the Clean Air Interstate Rule, and the
Clean Air Mercury Rule.
The revisions are prompted primarily by changes being implemented
by EPA's Clean Air Markets Division in its data systems in order to
utilize the latest modern technology for the submittal of data by
affected sources. Other revisions address issues that have been raised
during program implementation, fix specific inconsistencies in rule
provisions, or update sources incorporated by reference. These
revisions do not impose significant new requirements upon sources with
regard to monitoring or quality assurance activities.
DATES: This final rule is effective on January 24, 2008, for good cause
found as explained in this rule.
The incorporation by reference of certain publications listed in
the rule is approved by the Director of the Federal Register as of
January 24, 2008, for good cause found as explained in this rule.
ADDRESSES: The EPA has established a docket for this action under
Docket ID No. EPA-HQ-OAR-2005-0132. All documents in the docket are
listed in the http://www.regulations.gov index. Although listed in the index,
some information is not publicly available, e.g., CBI or other
information whose disclosure is restricted by statute. Certain other
material, such as copyrighted material, will be publicly available only
in hard copy. Publicly available docket materials are available either
electronically in http://www.regulations.gov or in hard copy at the Air and
Radiation Docket, EPA/DC, EPA West Building, EPA Headquarters Library,
Room 3334, 1301 Constitution Avenue, NW., Washington, DC. The Public
Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through
Friday, excluding legal holidays. The telephone number for the Public
Reading Room is (202) 566-1744, and the telephone number for the Air
and Radiation Docket is (202) 566-1742.
FOR FURTHER INFORMATION CONTACT: Matthew Boze, Clean Air Markets
Division, U.S. Environmental Protection Agency, Clean Air Markets
Division, MC 6204J, Ariel Rios Building, 1200 Pennsylvania Ave., NW.,
Washington, DC 20460, telephone (202) 343-9211, e-mail at
boze.matthew@epa.gov. Electronic copies of this document can be
accessed through the EPA Web site at: http://www.epa.gov/airmarkets.
SUPPLEMENTARY INFORMATION: Regulated Entities. Entities regulated by
this action primarily are fossil fuel-fired boilers, turbines, and
combined cycle units that serve generators that produce electricity,
generate steam, or cogenerate electricity and steam. Some trading
programs include process sources, such as process heaters or cement
kilns. Although Part 75 primarily regulates the electric utility
industry, certain State and Federal NOX mass emission
trading programs rely on subpart H of Part 75, and those programs may
include boilers, turbines, combined cycle, and certain process units
from other industries. Regulated categories and entities include:
----------------------------------------------------------------------------------------------------------------
Examples of potentially regulated
Category NAICS code industries
----------------------------------------------------------------------------------------------------------------
Industry................................ 221112 and others.......... Electric service providers Process
sources with large boilers, turbines,
combined cycle units, process heaters,
or cement kilns where emissions exhaust
through a stack.
----------------------------------------------------------------------------------------------------------------
This table is not intended to be exhaustive, but rather to provide
a guide for readers regarding entities likely to be regulated by this
action. This table lists the types of entities which EPA is now aware
could potentially be regulated by this action. Other types of entities
not listed in this table could also be regulated. To determine whether
your facility, company, business, organization, etc., is regulated by
this action, you should carefully examine the applicability provisions
in Sec. Sec. 72.6, 72.7, and 72.8 of title 40 of the Code of Federal
Regulations and in 40 CFR Parts 96 and 97. If you have questions
regarding the applicability of this action to a particular entity,
consult the person listed in the preceding FOR FURTHER INFORMATION
CONTACT section.
World Wide Web (WWW). In addition to being available in the docket,
an electronic copy of the final rule is also available on the WWW
through the Technology Transfer Network Web site (TTN Web). Following
signature, a copy of the rule will be posted on the TTN's policy and
guidance page for newly proposed or promulgated rules at http://www.epa.gov/ttn/oarpg.
The TTN provides information and technology
exchange in various areas of air pollution control.
Judicial Review. Under CAA section 307(b), judicial review of this
final action is available only by filing a petition for review in the
U.S. Court of Appeals for the District of Columbia Circuit on or before
March 24, 2008. Under CAA section 307(d)(7)(B), only those objections
to the final rule that were raised with specificity during the period
for public comment may be raised during judicial review. Moreover,
under CAA section 307(b)(2), the requirements established by today's
final rule may not be challenged separately in any civil or criminal
proceedings brought by EPA to enforce these requirements. Section
307(d)(7)(B) also provides a mechanism for the EPA to convene a
proceeding for reconsideration if the petitioner demonstrates that it
was impracticable to raise an objection during the public comment
period or if the grounds for such objection arose after the comment
period (but within the time for judicial review) and if the objection
is of central relevance to the rule. Any person seeking to make such a
demonstration to EPA should submit a Petition for Reconsideration,
clearly labeled as such, to the Office of the Administrator, U.S. EPA,
Room 3000, Ariel Rios Building, 1200 Pennsylvania Ave., Washington, DC
20460, with a copy to the Associate General Counsel for the Air and
Radiation Law Office, Office of General Counsel, Mail Code 2344A, U.S.
EPA, 1200 Pennsylvania Ave., NW., Washington, DC 20460.
Outline
I. Detailed Discussion of Rule Revisions
A. Rule Definitions
B. General Monitoring Provisions
C. Certification Requirements
[[Page 4313]]
D. Missing Data Substitution
E. Recordkeeping and Reporting
F. Subpart H (NOX Mass Emissions)
G. Subpart I (Hg Mass Emissions)
H. Appendix A
I. Appendix B
J. Appendix D
K. Appendix E
L. Appendix F
M. Appendix G
N. Appendix K
O. Other Rule Revisions
II. Statutory and Executive Order Reviews
A. Executive Order 12866: Regulatory Planning and Review
B. Paperwork Reduction Act
C. Regulatory Flexibility Act
D. Unfunded Mandates Reform Act
E. Executive Order: 13132: Federalism
F. Executive Order 13175: Consultation and Coordination With
Indian Tribal Governments
G. Executive Order 13045: Protection of Children From
Environmental Health and Safety Risks
H. Executive Order 13211: Actions That Significantly Affect
Energy Supply, Distribution, or Use
I. National Technology Transfer and Advancement Act
J. Executive Order 12898: Federal Actions To Address
Environmental Justice in Minority Populations and Low-Income
Populations
K. Congressional Review Act
L. Petition for Judicial Review
M. Determination Under Section 307(d)
I. Detailed Discussion of Rule Revisions
EPA is in the process of re-engineering the data systems associated
with the collection and processing of emissions, monitoring plan,
quality assurance, and certification data. The re-engineering project
includes the creation of a client tool, provided by EPA that sources
will use to evaluate and submit their Part 75 monitoring data. This
process change will enable sources to assess the quality of their data
prior to submitting the data using EPA established checking criteria.
The process will also allow sources to report their data directly to a
database. Having the data in a true database will allow the Agency to
implement and assess the program more efficiently and will streamline
access to the data. Also, this database structure will enable EPA to
implement process changes that will reduce the redundant reporting of
certain types of data. The re-engineered systems will be supported by a
new extensible markup language (XML) data format that will replace the
record type/column format currently used by EPA to collect electronic
data. EPA intends to transition existing sources to the new XML
electronic data report (XML-EDR) format during the 2008 reporting year.
For sources reporting in 2008 for the first time, the new XML-EDR
format should be used. All sources will be required to use the new
process beginning in 2009.
Therefore, EPA finds good cause to determine that the final rule is
effective on January 24, 2008. EPA normally issues final regulations
with at least a 30-day effective date after Federal Register
publication. However, this provision of the rule which pertains to the
re-engineering of the Clean Air Markets Division's data systems and to
implementation of the Clean Air Mercury Regulation (CAMR), must be
effective by January 1, 2008. Today's rule allows sources the option of
reporting emissions data in the new XML data reporting format in 2008,
one year before the use of XML becomes mandatory. The final rule
provides the necessary record keeping and reporting requirements to
support the XML format. Second, sources subject to CAMR are required to
install and certify continuous mercury (Hg) monitoring systems by
January 1, 2009. To meet this deadline, companies with multiple CAMR-
affected units will begin monitor certification testing in the first
quarter of 2008. As described in Sections I.C.3 and I.O.3., today's
rule adds two recently-published Hg test methods, i.e., Methods 30A and
30B, to Part 75 as alternatives to the Ontario Hydro Method. For many
sources, 30A and 30B will be the test methods of choice. Third, as
discussed in Section I.A., today's rule defers until January 1, 2010
the requirement for the calibration standards used to certify Hg
continuous emission monitoring systems (CEMS) under CAMR to be
traceable to the National Institute of Standards and Technology (NIST).
Fourth, for CAMR units that seek to qualify as low mass emitting units
under Sec. 75.81, Hg emission testing is required in 2008. As
discussed in Section G.2., today's rule adds considerable flexibility
to the way in which this testing is conducted, particularly for common
stack configurations and groups of identical units. The use of Methods
30A and 30B for this testing is also desirable. Absent this
determination of good cause, sources would not be able to begin
scheduled monitoring certification activities until the necessary
provisions of this rule became effective. A thirty day delay would
significantly decrease the overall amount of time available for
industry to comply with the certification deadline of January 1, 2009.
Such a delay could result in sources not being able to meet the
certification deadline, since industry would lose some of its ability
to spread utilization of various certification resources (i.e., test
teams, equipment, and vendor support) over the entire course of 2008.
For these reasons, EPA believes it has good cause to expedite the
effective date of this final rule.
A. Rule Definitions
Background
EPA proposed to add several new definitions to Part 72, including
definitions for: ``Long-term cold storage'' (to mean the complete
shutdown of a unit intended to last for at least two calendar years);
``EPA Protocol Gas Verification Program'' (to support the proposed
calibration gas audit program); ``Air Emission Testing Body (AETB)''
and ``Qualified Individual'' (to support the proposed stack tester
accreditation program).
EPA also proposed to modify the definitions of ``Capacity factor'',
``EPA protocol gas,'' and ``Excepted monitoring system'', and to remove
the definition of ``Calibration gas'' and related definitions
describing the various types of gas standards that are classified as
calibration gas.
Summary of Rule Changes
All of the proposed new and modified definitions have been
finalized without substantive changes. However, one commenter cautioned
that removing the definitions of the calibration gas standards from
Part 72 might have consequences that could necessitate further rule
revisions. In view of this, the Agency reconsidered these proposed
changes and the final rule retains all but one of the definitions. The
definition of ``Research gas material'' was found to be identical to
the definition of ``Research gas mixture'' and has been removed from
the rule.
Further, for consistency with Method 30A, the new instrumental
reference method for mercury (Hg) (which, as noted in sections I.C.3
and I.O.3 of this preamble has been added to the list of acceptable Hg
reference methods in Sec. 75.22), and in light of other changes in
today's rule related to the certification of Hg monitoring systems, EPA
is adding definitions of ``NIST traceable elemental Hg standards'' and
``NIST traceable source of oxidized Hg'' to Sec. 72.2. These
definitions pertain to Hg calibration gas standards and are deemed
necessary for implementation of the continuous monitoring requirements
of the Clean Air Mercury Regulation (CAMR).
Affected units under CAMR are required to install and certify Part
75-compliant Hg monitoring systems by January 1, 2009. To meet this
requirement, the vast majority of the
[[Page 4314]]
certification testing will be performed in 2008. When CAMR was first
proposed, only one reference test method (the Ontario Hydro (OH)
Method) was prescribed for the relative accuracy test audits (RATAs) of
the required Hg monitoring systems. However, the OH method is wet
chemistry-based, and is both difficult and expensive to perform. Also,
the laboratory analysis required to obtain the test results can take a
week or more, making the OH method incompatible with the Hg emissions
trading program described in the CAMR model rule.
In a cap and trade program, the RATA results must be known while
the test team is still on-site, so that any necessary corrective
actions can be taken and retesting performed without delay. With the OH
method, if the results of the lab analysis indicate a RATA failure, a
retest must be rescheduled and the Hg monitoring system is considered
out-of-control until a subsequent RATA is passed. This can result in an
extended missing data period and loss of Hg allowances.
Thus, it became apparent during the CAMR rulemaking that an
alternative to the OH method was needed. An instrumental Hg reference
method was put forth as the logical choice, because it would provide
real-time Hg concentration data, allowing the RATA results to be known
on the day of the test. When CAMR was published on May 18, 2005, EPA
stated its intention to ``propose and promulgate'' an instrumental Hg
reference method (see 70 FR 28636). In support of the final CAMR rule,
Hg monitoring provisions were added to Part 75. Among these was an
amendment to Sec. 75.22, allowing the use of either the OH method or
an ``instrumental reference method * * * subject to the approval of the
Administrator'' for the certification testing of Hg continuous
monitoring systems. Method 30A was published on September 7, 2007 in a
direct-final rulemaking, and became effective on November 6, 2007 (see
72 FR 51494). Method 30A represents the fulfillment of the Agency's
commitment to publish an instrumental reference method for Hg.
One of the most important Part 75 requirements for the
certification of Hg continuous emission monitoring systems (CEMS) is
that the concentrations of the elemental and oxidized Hg calibration
gas standards used for the 7-day calibration error tests, linearity
checks, and system integrity checks of the CEMS must be traceable to
the National Institute of Standards and Technology (NIST) (see Part 75,
Appendix A, Section 5.1.9). This NIST traceability requirement for Hg
standards is modeled after the NIST traceability requirements in
Section 5 of Appendix A for SO2, NOX, and diluent
gas (CO2 and O2) calibration gas standards.
For the SO2, NOX, CO2, and
O2 compressed gas standards used in Part 75 applications,
``NIST traceability'' means that the calibration gases have been
prepared according to the EPA-approved protocol cited in Section 5.1.4
of Appendix A. Further, Sec. 75.22(c)(1) requires NIST-traceable gas
standards to be used to calibrate the instrumental reference methods
used for relative accuracy testing of SO2, NOX,
CO2, and O2 CEMS (i.e., Methods 6C, 7E and 3A).
Prior to today's rulemaking, no NIST traceability protocols for Hg
calibration standards were referenced in Part 75. The new definitions
of ``NIST traceable elemental Hg standards'' and ``NIST traceable
source of oxidized Hg'' address this deficiency and cite the EPA
protocols that must be followed to ensure that the elemental and
oxidized Hg standards are traceable to NIST. However, these protocols,
which are referenced in Section 16.0 of Method 30A, are not yet fully
developed, and are not expected to be ready for use until the latter
part of 2008. A cooperative field demonstration program that will
include representatives from EPA, NIST, industry, equipment vendors,
and other key personnel is planned for the coming months, to gather the
data necessary to refine and finalize the traceability protocols. Once
these traceability protocols are finalized, they will be posted on the
Agency's Technology Transfer Network Web site (http://www.epa.gov/ttn/emc/) and on the Agency's Clean Air Markets Division Web site (http://
http://www.epa.gov/airmarkets/).
In view of this, EPA is temporarily deferring (until January 1,
2010) the requirement for elemental and oxidized Hg standards to be
NIST traceable. The deferral affects both initial certifications of the
CEMS and routine quality-assurance tests of the CEMS performed prior to
January 1, 2010. Note that only the NIST traceability requirement for
the Hg calibration standards is being waived, not the requirement to
perform the calibration error tests, linearity checks, and system
integrity checks of the Hg monitoring systems by January 1, 2009.
Beginning on January 1, 2010, all daily calibration error tests,
linearity checks, and system integrity checks of Hg CEMS must be
performed using NIST traceable elemental and oxidized Hg calibration
standards, as defined in Sec. 72.2. Section 5.1.9 of Appendix A to
Part 75 has been revised to reflect this. In view of this, EPA strongly
recommends that in 2009, all CAMR-affected sources should take the
necessary steps to ensure that the NIST traceability requirement is
met. In most cases, this will involve the certification of elemental
and oxidized Hg generators, according to the traceability protocols. If
a source elects to perform daily calibrations and/or linearity checks
using compressed gas cylinders instead of an elemental Hg generator,
the owner or operator will have to obtain cylinder gases that conform
to the EPA traceability protocol for gaseous calibration standards.
Finally, note that EPA is conditionally allowing Method 30A to be
used for Part 75 Hg emission testing and RATA applications prior to
finalization of the traceability protocols in section 16.0 of the
method. The condition is that interim traceability protocols are
developed and posted on the Agency's Technology Transfer Network Web
site (http://www.epa.gov/ttn/emc/), as ``broadly applicable alternative
test method approvals'' that will expire when the final protocols are
issued. EPA's authority to approve such test method alternatives is
described in 72 FR 4257, January 30, 2007.
EPA believes that a phased-in approach to NIST traceability is
appropriate and necessary, in light of the additional time needed to
finalize the traceability protocols and the time required for the
affected sources and equipment vendors to set up the necessary
infrastructure to implement the protocols. The Agency also believes
that this approach will not compromise the quality of the data for the
emissions trading program under CAMR, since in 2010, the first year in
which Hg emissions count against allowances held, NIST traceability of
the Hg calibration standards is mandatory.
B. General Monitoring Provisions
1. Update of Incorporation by Reference (Sec. 75.6)
Background
Section 75.6 identifies a number of methods and other standards
that are incorporated by reference into Part 75. This section includes
standards published by the American Society for Testing and Materials
(ASTM), the American Society of Mechanical Engineers (ASME), the
American National Standards Institute (ANSI), the Gas Processors
Association (GPA), and the American Petroleum Institute (API). EPA
proposed changes to Sec. 75.6 that would reflect the need to
incorporate
[[Page 4315]]
recent updates for many of the referenced standards. The proposed
revisions would recognize or adhere to these newer standards by
updating references for the standards listed in Sec. Sec. 75.6(a)
through 75.6(f). Additionally, new Sec. Sec. 75.6(a)(45) through
75.6(a)(48) and 75.6(f)(4) would incorporate by reference additional
ASTM and API standards that are relevant to Part 75 implementation.
Summary of Rule Changes
The updates and additions to Sec. 75.6 have been finalized as
proposed. One commenter requested that an additional ASTM method for
analyzing the sulfur content of low-sulfur fuel oil, i.e., ASTM D5453-
06, ``Standard Test Method for Determination of Total Sulfur in Light
Hydrocarbons, Spark Ignition Engine Fuel, Diesel Engine Fuel, and
Engine Oil by Ultraviolet Fluorescence'', be added to the list of
acceptable methods in Sec. 75.6. This method has been incorporated by
reference as Sec. 75.6(a)(49) and has been added to section 2.2.5 of
Appendix D.
2. Default Emission Rates for Low Mass Emissions (LME) Units
Background
EPA proposed to allow LME units to use site-specific default
SO2 emission rates for fuel oil combustion, in lieu of using
the ``generic'' default SO2 emission rates specified in
Table LM-1 of Sec. 75.19. To use this option, a federally enforceable
permit condition would have to be in place for the unit, limiting the
sulfur content of the oil. This revision, if made, would allow more
representative, yet still conservatively high, SO2 emissions
data to be reported from oil-burning LME units. As proposed, the site-
specific default SO2 emission rate would be calculated using
an equation from EPA publication AP-42. The sulfur content used in the
calculations would be the maximum weight percent sulfur allowed by the
federally-enforceable permit. Sources choosing to implement this option
would be required to perform periodic oil sampling using one of the
four methodologies described in Section 2.2 of Appendix D to Part 75,
and would be required to keep records documenting the sulfur content of
the fuel.
The Agency also proposed to revise Sec. 75.19(c)(1)(iv)(G) to
clarify that fuel-and-unit-specific default NOX emission
rates for LME units may be determined using data from a Continuous
Emissions Monitoring System (CEMS) that has been quality-assured
according to either Appendix B of Part 75 or Appendix F of Part 60, or
comparably quality-assured under a State CEMS program. Lastly, the
Agency proposed technical revisions to the Equations LM-5 and LM-6
changing the units of rate to units of measure to make the equations
correct as units of rate cannot technically be summed.
Summary of Rule Changes
Commenters were generally supportive of the proposed revisions to
Sec. 75.19, and they have been finalized with only one substantive
change. EPA has incorporated one commenter's suggestion not to restrict
the allowable fuel oil sampling options to those described in Appendix
D. The final rule allows the use of other consensus standard fuel
sampling methods (e.g., ASTM, API, etc.) specified in applicable State
or Federal regulations or in the unit's operating permit, to determine
the sulfur content of the oil.
Another commenter requested that EPA go beyond its proposal for
SO2 and consider providing a similar, more reasonable site-
specific alternative to reporting the generic NOX emission
rates in Table LM-2. Specifically, the commenter suggested that for
units with very low annual capacity factors, the Agency should waive
the testing requirements of Sec. Sec. 75.19(c)(1)(iv) and allow
emission test data that was generated more than 5 years ago (e.g., from
a Part 60 performance test) to be used to determine fuel-specific
default NOX emission rates. The commenter asserted that the
cost of additional testing could impose a financial burden on smaller
affected sources. After careful consideration, EPA decided against
allowing infrequently-operated units to use emission test data older
than 5 years for Part 75 reporting. However, Sec. 75.19(c)(1)(iv)(I)
has been amended to provide reduced emission testing requirements for
very low capacity factor LME units. The final rule allows single-load
testing, between 75 and 100 percent of maximum load, to be performed
(both for the initial Appendix E testing and for retests) if, for the 3
years prior to the year of the test, the unit's average capacity factor
was 2.5 percent or less and did not exceed 4.0 percent in any of those
three years. Alternatively, for combustion turbines, the emission test
may be done at the maximum attainable load corresponding to the season
of the year in which the test is performed. For a group of identical
units, the single-load testing option may be used for any unit(s) in
the group that meet the very low capacity factor requirements. For a
more detailed discussion of this issue, refer to section 2.3.2 of the
Response to Comments (RTC) document.
3. Default Moisture Value for Natural Gas
Background
EPA proposed to allow gas-fired boilers equipped with CEMS to use
default moisture values in lieu of continuously monitoring the stack
gas moisture content. Two conservative default values were proposed:
14.0% H2O under Sec. 75.11(b), and 18.0% H2O under Sec.
75.12(b). The Agency also proposed that the higher default value would
apply only when Equation 19-3, 19-4, or 19-8 (from Method 19 in
appendix A-7 to part 60 of this chapter) is used to determine the
NOX emission rate. The proposed default values represent the
10th and 90th percentile values from two sets of supplemental moisture
data provided to the Agency, which is consistent with the approach that
the Agency has used in responding to past petitions under Sec. 75.66
for site-specific default moisture values.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized.
4. Expanded Use of Equation F-23
Background
EPA proposed to revise Sec. 75.11(e)(1) to remove the current
restrictions on the use of Equation F-23 to determine the
SO2 mass emission rate, by allowing Equation F-23 to be used
whether or not the unit has an SO2 monitor and to expand its
use to fuels other than natural gas. The proposal would allow Equation
F-23 to be used for any gaseous fuel that qualifies for a default
SO2 emission rate under Section 2.3.6(b) of Appendix D.
Further, Equation F-23 could be used for the combustion of liquid and
solid fuels that meet the definition of ``very low sulfur fuel'' in
Sec. 72.2, if a petition for a fuel-specific default SO2
emission rate is submitted to the Administrator under Sec. 75.66 and
the Administrator approves the petition. Under the proposed rule,
petitions would also be accepted for the combustion of mixtures of
these fuels and for the co-firing of these fuels with gaseous fuel.
Summary of Rule Changes
Commenters were supportive of the expanded use of Equation F-23 and
the revisions to Sec. 75.11(e) and corresponding changes to section 7
of Appendix F have been finalized as proposed.
[[Page 4316]]
5. Calculation of NOX Emission Rate--LME Units
Background
EPA proposed to re-title Sec. 75.19(c)(4)(ii) as ``NOX
mass emissions and NOX emission rate'' and to add a new
subparagraph (D) to Sec. 75.19 (c)(4)(ii), providing instructions for
determining quarterly and cumulative NOX emission rates for
a LME unit. The NOX emission rate for each hour (lb/mmBtu)
would simply be the appropriate generic or unit-specific default
NOX emission rate defined in the monitoring plan for the
type of fuel being combusted and (if applicable) the NOX
emission control status. Then, the Agency proposed that the quarterly
NOX emission rate would be determined by averaging all of
the hourly NOX emission rates and the cumulative (year-to-
date) NOX emission rate would be the arithmetic average of
the quarterly values.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and the revisions to Sec. 75.19(c)(4)(ii) have been finalized as
proposed.
6. LME Units--Scope of Applicability
Background
EPA proposed to revise Sec. 75.19(a)(1) to clarify that the low
mass emissions (LME) methodology is a stand-alone alternative to a CEMS
and/or the ``excepted'' monitoring methodologies in Appendices D, E,
and G. In other words, if a unit qualifies for LME status, the owner or
operator is required either to use the LME methodology for all
parameters or not to use the method at all. No mixing-and-matching of
other monitoring methodologies with LME is permitted. Parallel
revisions to Sec. Sec. 75.11(d)(3), 75.12(e)(3), and 75.13(d)(3),
consistent with the changes to Sec. 75.19(a)(1), were also proposed to
clarify the Agency's intent.
Summary of Rule Changes
No adverse comments were received on the proposed changes and they
have been finalized.
7. Use of Maximum Controlled NOX Emission Rate When Using
Bypass Stacks
Background
Revisions to Sec. 75.17(d)(2) were proposed that would allow a
maximum controlled NOX emission rate (MCR) to be reported
instead of the maximum potential NOX emission rate (MER)
whenever an unmonitored bypass stack is used, provided that the add-on
controls are not bypassed and are documented to be operating properly.
For example, for a coal-fired unit equipped with FGD and SCR add-on
emission controls, if the SCR is documented to be working during an FGD
malfunction and the effluent gases are routed through an unmonitored
bypass stack after passing through the SCR, then the MCR, rather than
the MER, would be the more appropriate NOX emission rate to
report for the bypass hour(s). Documentation of proper add-on control
operation for such hours of operation would be required as described in
Sec. 75.34(d). The MCR would be calculated in a manner similar to the
calculation of the MER, except that the maximum expected NOX
concentration (MEC) would be used instead of the maximum potential
NOX concentration (MPC).
Summary of Rule Changes
Commenters were generally supportive of the proposed rule changes
and they have been finalized. One commenter recommended that parallel
language be added to Sec. 75.72(c)(3), to cover non-Acid Rain Program
units that are subject to the NOX mass emissions monitoring
provisions of Subpart H. EPA agrees with this comment and has added the
necessary language to Sec. 75.72(c)(3).
C. Certification Requirements
1. Alternative Monitoring System Certification
Background
EPA proposed to delete Sec. Sec. 75.20(f)(1) and (2) from the
rule, thereby removing the requirement for the Administrator to publish
each request for certification of an alternative monitoring system in
the Federal Register, with an associated 60-day public comment period.
This rule provision is considered unnecessary, in view of the Agency's
authority under Subpart E to approve alternative monitoring systems and
the rigorous requirements in Sec. Sec. 75.40 through 75.48 that
alternative monitoring systems must meet in order to be certified.
Summary of Rule Changes
Commenters were supportive of the proposed amendments to Sec.
75.20(f), and they have been finalized.
2. Part 60 Reference Test Methods
Background
On May 15, 2006, EPA promulgated final revisions to EPA reference
test methods 6C, 7E, and 3A, which are found in Appendix A of 40 CFR
Part 60. (See 71 FR 28082, May 15, 2006). These test methods are
prescribed for Part 75 emission testing and RATAs. Three new testing
options that were added to the methods were deemed unacceptable for use
under Part 75. These include:
(1) Section 7.1 of revised EPA Method 7E, allowing for custom
calibration gas concentrations to be produced by diluting EPA protocol
gases, in accordance with Method 205 in Appendix M of 40 CFR Part 51.
(2) Section 8.4 of revised EPA Method 7E, allowing the use of a
multi-hole ``rake'' probe to satisfy the multipoint traverse
requirement of the method.
(3) Section 8.6 of revised EPA Method 7E, allowing for the use of
``dynamic spiking'' as an alternative to the interference and system
bias checks of the method.
Although revised Method 7E states that for use under Part 75 the
three options above require approval by the Administrator, EPA proposed
to add similar language to Sec. 75.22(a)(5) to reinforce its position
regarding these testing alternatives.
Summary of Rule Changes
No adverse comments were received on the proposed amendments to
Sec. 75.22(a)(5) and they have been finalized. However, one commenter
brought to EPA's attention another revision to the Part 60 reference
methods that impacts Part 75. EPA Method 20 was also revised on May 15,
2006. Method 20 has been the NOX emission test method
prescribed for combustion turbines (CTs) in section 2.1.2.2 of Appendix
E. Method 20 has also been used to determine fuel-specific
NOX emission rates for combustion turbines that qualify as
low mass emissions (LME) units under Sec. 75.19.
The original Method 20 required testing at 8 sampling points per
run, with typical run times averaging about 15 to 20 minutes. However,
the revised Method 20 no longer specifies the minimum number of test
points per run, but rather requires sampling point selection to be done
according to Method 7E. Revised Method 7E requires 12 traverse points
for an emission test run (which would suffice for Appendix E testing),
but the method also allows the results of stratification testing to be
used to justify using three or, in some cases, one sample point
instead. This raises questions about the required length of an Appendix
E test run. For instance, if testing were required at only one point,
each Appendix E test run would be reduced from 15-20 minutes to as
little as 2 minutes (depending on the system response time). The
commenter stated that such short sampling runs seem inadequate to
[[Page 4317]]
develop a substantial correlation curve for emission reporting. The
commenter recommended that EPA modify Appendix E or Method 20 and
either set a minimum run time of 20 minutes (providing an hour of data
at each load) or specify a minimum number of sampling points for an
Appendix E test of a CT.
EPA has incorporated the commenter's recommendations into Part 75.
First, Sec. 75.22(a)(5) has been amended to prohibit the use of Method
7E to determine the required number of sample points for the emission
testing of a combustion turbine. Section 75.22(a)(5)(ii) requires the
sample points to be determined according to section 2.1.2.2 of Appendix
E, instead. Second, for the emission test of a CT, section 2.1.2.2 of
Appendix E has been revised to require a minimum of 12 test points per
run, located according to EPA Method 1. Third, amendments have been
made to Sec. 75.22(a)(6), Sec. 75.19(c)(1)(iv)(A), section 6.5.10 of
Appendix A, and sections 2.1.2.2 and 2.1.2.3 of Appendix E, to remove
all references to EPA Method 20 from Part 75. Fourth, for the testing
of an Appendix E boiler, the text of section 2.1.2.1 of Appendix E has
been revised to require 12 traverse points per run, making it
consistent with revised section 2.1.2.2 (note that this is not a new
requirement--section 2.1.2.1 has always required 12 test points,
located according to section 8.3.1 of Method 3, and that section refers
back to Method 1). Finally, in section 2.1.2.3 of Appendix E, the
references to the measurement system response time in section 5.5 of
Method 20 (which section no longer exists) have been replaced with
references to the response time provisions in sections 8.2.5 and 8.2.6
of Method 7E. Appendix E tests performed on CTs prior to the effective
date of these amendments are grandfathered from the revised test point
location requirements.
3. Mercury Reference Methods
Background
EPA proposed to add an alternative relative deviation (RD)
specification for the results of mercury (Hg) emission data collected
with paired Ontario Hydro (OH) reference method sampling trains. The
principal RD specification in Sec. 75.22(a)(7) is 10 percent. However,
this acceptance criterion may be too stringent for sources with low Hg
emissions. Therefore, for average Hg concentrations of 1.0 [mu]g/m\3\
or less, EPA proposed an alternative RD specification of 20 percent.
This is consistent with the acceptance criteria for data from paired OH
trains, as specified in Performance Specification 12A in Appendix B of
40 CFR Part 60.
EPA also proposed amendments to Sec. Sec. 75.22(a)(7),
75.59(a)(7), 75.81(c)(1), and to sections 6.5.10 and 7.6.1 of Appendix
A, allowing EPA Method 29 (back-half impinger catch, only) to be used
as an alternative to the OH method, both for RATA testing and for
periodic emission testing of units with low Hg mass emissions (< =29 lb/
yr). Two caveats on the use of Method 29 were proposed. First, sources
electing to use Method 29 (which is similar to the OH method, but
somewhat simpler and more familiar to stack testers) would be required
to use paired sampling trains (i.e., two trains sampling the source
effluent simultaneously), and the RD specifications in Sec.
75.22(a)(7) would have to be met for each run. Second, certain
analytical and quality assurance (QA) procedures in the OH method (ASTM
D6784-02) would have to be followed instead of the corresponding
procedures in Method 29 (because the analytical and quality assurance/
quality control (QA/QC) requirements of the OH method are more detailed
and rigorous than those in Method 29), and testers could opt to follow
several of the sample recovery and preparation procedures in the OH
method instead of the Method 29 procedures.
Finally, the Agency solicited comment on the use of sorbent traps
for reference method testing. Members of the regulated community had
expressed an interest in using portable sorbent trap monitoring systems
for Hg reference method testing, as an alternative to the OH method.
EPA proposed to accommodate a possible future sorbent-based reference
method by adding language to Sec. 75.22(a)(7) that would allow an
``other suitable'' reference method approved by the Administrator to be
used for Hg emission testing and RATAs.
Summary of Rule Changes
Commenters were generally supportive of the proposed amendments
that would add Method 29 as an alternative Hg reference method, and
those provisions have been finalized without substantive change. One
commenter objected to the requirement to use paired sampling trains for
OH and Method 29 tests, asserting that this adds to the cost of testing
and may result in significant numbers of test runs being discarded.
However, EPA does not agree with the commenter. The Agency believes
rather that paired sampling trains provide added assurance of data
quality when these test methods are used. The decision to require
paired trains for the OH method was made during the rulemaking that led
to publication of the Clean Air Mercury Regulation (CAMR) (see 70 FR
28636-28639, May 18, 2005).
Two commenters supported the proposed 20 percent alternative RD
specification for low emitters, and that provision has been finalized.
However, one of the commenters noted that even a 20 percent RD
specification may be too stringent for extremely low Hg concentrations.
EPA agrees that when Hg concentrations are exceptionally low (0.1
[mu]g/m\3\ or less), the 20 percent RD specification may be difficult
to meet. Therefore, the final rule adds a third tier to the RD
specifications in Sec. 75.22. The paired train agreement is also
considered to be acceptable if the absolute difference between the two
measured Hg concentrations does not exceed 0.03 [mu]g/m\3\.
Several commenters strongly supported the proposal to allow the use
of a sorbent-based reference method for Hg emission testing and for the
RATAs of Hg monitoring systems. Since publication of the proposed rule,
a great deal of progress has been made in this area. First, EPA
conducted a Method 301 analysis of available data comparing sorbent
trap sampling to the OH method. The results of this analysis showed
that a sorbent-based sampling method can be a viable alternative
reference method. Second, EPA drafted ``Method 30B'', a reference
method that uses iodated carbon traps to measure vapor phase Hg
emissions. Finally, as part of a direct final rulemaking, Method 30B
was published on September 7, 2007 (see 72 FR 51494-51531), along with
Method 30A, an instrumental Hg reference method. Today's final rule
allows both Methods 30A and 30B to be used.
D. Missing Data Substitution
1. Block Versus Step-Wise Approach
Background
Historically, EPA's policy has required sources to use a ``block''
approach for CEMS missing data substitution. The percent monitor data
availability (PMA) at the end of the missing data period has been used
to determine which mathematical algorithm applies, and the substitute
data value or values prescribed by that one algorithm have been
reported for each hour of the missing data period.
However, EPA has recently reconsidered and revised its missing
substitution data policy, to allow sources to apply the missing data
algorithms in a stepwise manner instead of using the block approach.
Under the
[[Page 4318]]
stepwise methodology, the various missing data algorithms are applied
sequentially. That is, the least conservative algorithm is applied to
the missing data hours until the PMA drops below 95%. Then, the next
algorithm is applied until the PMA has dropped below 90%, and so on.
Since Part 75 is not clear about which of the two methods should be
used for missing data substitution, EPA proposed to amend Sec. Sec.
75.33 and 75.32(b), to clarify that the stepwise, hour-by-hour method
is the preferred one, and that use of that method would be required for
all CEMS data recorded on and after January 1, 2009, and for any CEMS
data recorded in XML-format during the transition year of 2008.
Summary of Rule Changes
Commenters unanimously supported the proposal to adopt stepwise
missing data substitution and the proposed amendments to Sec. Sec.
75.32 and 75.33 have been finalized.
2. Substitute Data Values for Controlled Units
Background
For units with add-on emission controls, when the PMA for
SO2 or NOX is below 90.0 percent, Sec.
75.34(a)(3) has historically allowed the designated representative (DR)
to petition the Administrator under Sec. 75.66 for permission to
report the maximum controlled concentration or emission rate recorded
in a specified lookback period instead of reporting the maximum value
recorded in that lookback period, for each missing data hour in which
the add-on controls are documented to be operating properly. After more
than ten years of implementing the Acid Rain Program, EPA no longer
believes that such special petitions are necessary, because sources
with add-on controls are required to implement a quality assurance/
quality control (QA/QC) program that includes the recording of
parametric data to document the hourly operating status of the emission
controls. This parametric information must be made available to
inspectors and auditors upon request. Therefore, any claim that the
emission controls were operating properly during a particular missing
data period can be easily verified through the audit process.
In view of this, the Agency proposed to remove from Sec.
75.34(a)(3) and Sec. 75.66(f) the requirement to petition the
Administrator to use the maximum controlled SO2 or
NOX concentration (or maximum controlled NOX
emission rate) from the applicable lookback period. The proposed
revisions would simply allow the maximum controlled values to be
reported whenever parametric data are available to document that the
emission controls are operating properly. The proposed rule would
further clarify that this reporting option applies only to the third
missing data tier, when the PMA is greater than or equal to 80.0
percent, but less than 90.0 percent.
EPA also proposed to add a new paragraph (a)(5) to Sec. 75.34,
which would allow units with add-on emission controls to report
alternative substitute data values for missing data periods in the
fourth missing data tier, when the PMA is below 80.0 percent. Proposed
Sec. 75.34(a)(5) would allow the owner or operator to replace the
maximum potential SO2 or NOX concentration (MPC)
or the maximum potential NOX emission rate (MER) with a less
conservative substitute data value, for missing data hours where
parametric data, (as described in Sec. Sec. 75.34(d) and 75.58(b)) are
available to verify proper operation of the add-on controls.
Specifically, for SO2 and NOX concentration, the
replacement value for the MPC would be the greater of: (a) The maximum
expected concentration (MEC); or (b) 1.25 times the maximum controlled
value in the standard missing data lookback period. For NOX
emission rate, the replacement value for the MER would be the greater
of: (a) The maximum controlled NOX emission rate (MCR); or
(b) 1.25 times the maximum controlled value in the standard missing
data lookback period. The NOX MCR would be calculated in the
same manner as the NOX MER, except that the MEC, rather than
the MPC, would be used in the calculation. The proposed alternative
data substitution methodology in Sec. 75.34(a)(5) would ensure that
the substitute data values for the fourth missing data tier are always
higher than the corresponding substitute data values for the third
tier.
Finally, EPA proposed to revise Sec. 75.38(c) to extend the
alternative missing data options for the third and fourth tiers to
mercury (Hg) concentration, and Sec. 75.58(b)(3) would be revised to
be consistent with the proposed revisions to Sec. Sec. 75.34(a)(3),
75.34(a)(5), and 75.38(c).
Summary of Rule Changes
Comments on the proposed alternative missing data substitution
values for controlled units were generally supportive and these
provisions have been finalized. Two commenters requested that parallel
language be added to Sec. 75.72(c)(3), to extend the use of the new
missing data provisions to ozone season-only reporters. Another
commenter asked EPA to clarify that the MCR may be implemented on a
fuel-specific basis. EPA has incorporated both of these suggestions in
the final rule. Two other commenters suggested that, for common stack
configurations, EPA should allow the substitute data values to be
apportioned or prorated in some way instead of requiring maximum
potential values to be reported, in cases where the emission controls
installed on some of the units sharing the stack are documented to be
operating properly, but such documentation cannot be provided for the
controls on the other units. The Agency believes that this approach
would unnecessarily complicate the missing data substitution process
and would provide no assurance that emissions are not being
underestimated. Therefore, this suggestion was not incorporated in the
final rule.
3. Substitute Data Values for Hg
Background
EPA proposed to revise the Hg missing data procedures. First, for
Hg CEMS, the text of Sec. 75.38(a) would be amended to clarify that
the PMA ``trigger conditions'' for Hg monitoring systems are different
from the trigger conditions for all other parameters. For all
parameters except Hg, the trigger points that define the boundaries of
the four missing data tiers are 95 percent, 90 percent, and 80 percent
PMA. However, for Hg the corresponding trigger points are 90 percent,
80 percent and 70 percent, respectively.
Second, EPA proposed to completely revise the missing data
provisions in Sec. 75.39 for sorbent trap monitoring systems, to make
them the same as for Hg CEMS, so that. the initial missing data
procedures of Sec. 75.31(b) and the standard Hg missing data
provisions of Sec. 75.38 would be followed for sorbent trap systems.
EPA believes that this proposed missing data approach greatly
simplifies the missing data substitution process for Hg monitoring
systems. The hourly Hg concentration data stream from a sorbent trap
system will look essentially the same as the data stream from a CEMS,
except that the Hg concentration will ``flat-line'' (i.e., will not
change) during each data collection period. Therefore, under the
proposal, when the owner or operator elects to use a primary Hg CEMS
and a backup sorbent trap system (or vice-versa), the appropriate
substitute data values would be derived from a lookback through the
previous 720 hours of quality-assured data, irrespective of
[[Page 4319]]
whether they were from the primary monitoring system or from the backup
system.
Summary of Rule Changes
Commenters were supportive of the proposed changes to the sorbent
trap missing data procedures in Sec. 75.39, and these provisions have
been finalized.
4. Correction of Cross-References
Background
For sources that report emissions data on an ozone season-only
basis, EPA proposed to revise Sec. 75.74(c)(3)(xi) and (c)(3)(xii) by
replacing references to specific missing data sections with more
general references to the entire block of CEMS missing data sections,
i.e., Sec. Sec. 75.31 through 75.37.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
E. Recordkeeping and Reporting
Background
To accommodate its new, re-engineered XML reporting format, which
will replace the current electronic data reporting (EDR) format in
2009, EPA proposed to revise the monitoring plan recordkeeping
requirements in Sec. 75.53, with corresponding revisions to Sec.
75.73(c)(3) (for sources reporting NOX mass emissions under
Subpart H) and to Sec. 75.84 (for sources reporting Hg mass emissions
under Subpart I).
EPA proposed to add two new paragraphs, (g) and (h), to Sec.
75.53, which describe the required monitoring plan data elements in
EPA's re-engineered XML data structure. Under this proposal, the
provisions of paragraphs (g) and (h) would be followed instead of the
existing recordkeeping requirements of paragraphs (e) and (f), on and
after January 1, 2009. In 2008, sources would be allowed to choose
between the EDR format and XML, but new sources reporting for the first
time in 2008 would be strongly encouraged to use the XML format.
Included among the proposed monitoring plan changes would be mandatory
recording and reporting of the key rectangular duct wall effects data
elements using these record types. The proposed requirements to record
and report the results of wall effects adjustment factor (WAF)
determinations in the monitoring plan are found in Sec. Sec. 75.53 (e)
and (g) and in Sec. 75.64.
EPA also proposed to make a series of modifications to Sec. Sec.
75.58 and 75.59 to support the new XML data structure. The proposed
changes to the monitoring plan and recordkeeping sections were
presented, section-by-section, in Tables 1, 2, and 3 in the preamble to
the August 22, 2006 proposed rule.
Summary of Rule Changes
No significant adverse comments were received on the proposed
changes and they have been finalized.
1. Other Reporting Issues
a. Long-Term Cold Storage and Deferred Units
Background
EPA proposed changes to Part 75 to clarify the meaning of the term
``long-term cold storage (LTCS)'', found in Sec. 75.4(d). First, a
proposed definition of long-term cold storage would be added to Sec.
72.2. LTCS would mean that the unit has been completely shut down and
placed in storage and that the shutdown is intended to last for an
extended period of time (at least two calendar years). Second, the
Agency proposed to add a new paragraph, (a)(7), to Sec. 75.61,
requiring the owner or operator to provide notifications when a unit is
placed in LTCS and when the unit re-commences operation. Third,
modifications to Sec. 75.20(b) were proposed, requiring
recertification of all monitoring systems when a unit re-commences
operations after a period of long-term cold storage. If a source
claiming LTCS status re-commenced operation sooner than two years after
being placed in LTCS, the notification and recertification requirements
would apply. Fourth, the proposed rule would exempt a unit in LTCS from
quarterly emissions reporting under Sec. 75.64 until the unit
recommences operation. Parallel LTCS rule provisions and appropriate
cross-references regarding quarterly reporting requirements for Subpart
H and Subpart I units would be added to Sec. Sec. 75.73(f)(1) and
75.84(f)(1), respectively, for consistency.
EPA also proposed to revise the provisions of Sec. Sec. 75.4(d)
and 75.61(a)(3) pertaining to ``deferred'' units, i.e., units for which
a planned or unplanned outage prevents the required continuous
monitoring systems from being certified by the compliance date. The
proposed revisions would broaden the scope of Sec. 75.4(d) beyond the
Acid Rain Program, to include units in State or Federal pollutant mass
emissions reduction programs that adopt the monitoring and reporting
provisions of Part 75. Examples of such programs include the Clean Air
Interstate Regulation (CAIR), which is scheduled to begin in 2008 and
the Clean Air Mercury Regulation (CAMR), which goes into effect in
2009. The proposed revisions to Sec. Sec. 75.4(d) and 75.61(a)(3) were
deemed necessary because the CAIR and CAMR rules do not address
deferred units.
The proposed revisions to Sec. 75.4(d) would require the owner or
operator of a deferred unit to provide notice of unit shutdown and
recommencement of commercial operation, either according to Sec.
75.61(a)(3) (for planned shutdowns such as scheduled maintenance
outages and for unplanned, forced unit outages) or Sec. 75.61(a)(7)
(for units in long-term cold storage). For all of these circumstances
involving deferred units, EPA proposed that the Part 75 continuous
monitoring systems would have to be certified within 90 unit operating
days or 180 calendar days (whichever comes first) of the date that the
unit recommences commercial operation. In the time interval between the
unit re-start and the completion of the required certification tests,
the owner or operator would be required to report emissions data, using
either: (1) Maximum potential values; (2) the conditional data
validation procedures of Sec. 75.20(b)(3); (3) EPA reference methods;
or (4) another procedure approved by petition to the Administrator
under Sec. 75.66. Finally, the Agency proposed to revise the
notification requirements of Sec. 75.61(a)(3) to be consistent with
the proposed changes to Sec. 75.4(d).
Summary of Rule Changes
Commenters were generally supportive of the proposed long-term cold
storage provisions, requesting only minor clarifications. These
provisions have been finalized with no substantive changes. One
commenter encouraged EPA to adopt the proposed amendments to broaden
the scope of Sec. 75.4(d), to ensure that deferred units under
programs such as CAIR and CAMR are provided with a reasonable window of
time in which to certify the required monitoring systems, when the
units resume operation. EPA has finalized these amendments to Sec.
75.4(d), as proposed.
b. Notice of Initial Certification Deadline
Background
EPA proposed to add a new paragraph (a)(8) to Sec. 75.61, to
require new and newly affected sources to notify EPA when the
monitoring system certification deadline is reached. Depending on the
program(s) to which the unit is subject, this date will always be a
particular number of calendar days or unit operating days after a unit
either:
[[Page 4320]]
(a) Commences commercial operation; (b) commences operation; or (c)
becomes an affected unit. For Acid Rain Program sources, the Agency
must know this date to correctly assess when to begin counting
emissions against allowances pursuant to Sec. 72.9. Knowing this date
also confirms that the monitoring systems either have or have not been
certified by the legal deadline.
Summary of Rule Changes
One commenter asserted that the requirement for sources to submit
to EPA a notification of the deadline for initial monitoring system
certification is unnecessarily burdensome and should not be
incorporated into Part 75. Another commenter requested that the
information be reported in the electronic monitoring plan, rather than
requiring a separate notification. EPA does not agree that reporting
this information will be burdensome or that it is appropriate to report
the date of the initial certification deadline in the electronic
monitoring plan. Rather, this date is an essential data element that
will be managed using the web-based CAMD Business System (CBS).
Therefore, the notification requirement can be met electronically using
the CBS. In view of this, the amendment to Sec. 75.61 has been
finalized, as proposed.
c. Monitoring Plan Submittal Deadline
Background
EPA proposed to amend Sec. 75.62(a) by changing the submittal
deadline for the initial monitoring plan for new and newly-affected
units from 45 days to 21 days prior to the initial certification
testing, in order to synchronize the initial monitoring plan submittal
with the initial test notice. Corresponding changes to Subpart H (Sec.
75.73(e)) and to Subpart I (Sec. 75.84(e)) were proposed, for
consistency.
EPA also proposed to remove the requirement from Sec. 75.62(a)(1)
that the electronic monitoring plan must be submitted ``in each
electronic quarterly report''. Rather, inclusion of the monitoring plan
in the report would be optional, and monitoring plan updates would be
made either prior to or concurrent with (but not later than) the date
of submission of the quarterly report. These proposed revisions would
allow sources to maintain their monitoring plan information separate
from the quarterly report, but this option would only be available to
sources reporting in the new XML format under the re-engineered data
submission process.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
d. EPA Form 7610-14
Background
EPA proposed to amend Sec. Sec. 75.63(a)(1) and (a)(2), to remove
the requirement to submit hardcopy EPA form 7610-14 along with every
certification or recertification application. Significant upgrades to
EPA's data systems have been made in recent years, and Form 7610-14 is
no longer needed to process these applications.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
e. LME Applications
Background
EPA proposed to remove the requirement from Sec.
75.63(a)(1)(ii)(A) for a hardcopy LME certification application to be
submitted to the Administrator. The proposal would require only the
electronic portion of the application, including the monitoring plan
and LME qualification records, to be sent to EPA's Clean Air Markets
Division. The hardcopy portion of the LME application would be sent to
the State and to the EPA Regional Office.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
f. Reporting Test Data for Diagnostic Events
Background
EPA proposed to revise Sec. 75.63(a)(2)(iii) to make the reporting
of the results of diagnostic tests more flexible. Rather than requiring
these test results to be reported in the electronic quarterly report
for the quarter in which the tests are performed, they could either be
submitted prior to or concurrent with that quarterly report. However,
this proposed flexibility in the reporting of diagnostic test results
would only be available to sources reporting in the new XML format
under the re-engineered data submission process.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
g. Modifications to Sec. 75.64
Background
As part of its data systems re-engineering effort, EPA proposed to
revise Sec. 75.64(a) to describe the transition from the existing EDR
reporting requirements to the reporting requirements of the new XML
format. The Agency proposed to renumber several paragraphs, to replace
paragraphs (a)(1) and (a)(2) with new paragraphs (a)(3) through (a)(7),
and to remove existing paragraph (a)(8).
Summary of Rule Changes
No adverse comments were received on these proposed rule changes.
These amendments to Sec. 75.64(a) have been finalized, as proposed.
h. Steam Load Reporting
Background
EPA proposed to add a third option to Part 75 for reporting load
data in units of mmBtu/hr of steam thermal output. This option is
needed to accommodate emissions trading programs in which allowance
allocations are made on an electrical or thermal output basis, rather
than a heat input basis. The Agency proposed to add text to several
sections in the main body of Part 75 and to the Appendices, to
accommodate the new reporting option.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
i. Test Notification Requirements--Hg Low Mass Emission Units
Background
Section 75.61(a)(5) requires the owner or operator or the
designated representative to provide 21-day advance notice for various
periodic quality-assurance tests, including the semiannual or annual
relative accuracy tests of CEMS, and for the re-tests of Appendix E
peaking units and low mass emissions (LME) units. Test notices must be
provided to the Administrator, to the appropriate EPA Regional Office
and to the State or local agency (unless a particular agency issues a
waiver from the requirement).
Under Subpart I of Part 75, certain low-emitting units covered by
the Clean Air Mercury Regulation (CAMR) may qualify under Sec. Sec.
75.81(b) through (d) to perform periodic (semiannual or annual) Hg
emission testing in lieu of operating and maintaining continuous Hg
monitoring systems. EPA proposed to expand the notification
requirements of Sec. 75.61(a)(5) and to add
[[Page 4321]]
corresponding introductory text to Sec. 75.61(a)(1), requiring the
owner or operator or the designated representative to provide at least
21 days notice of the scheduled dates of these periodic Hg emission
tests.
Summary of Rule Changes
No adverse comments were received on this proposed rule change and
this test notification requirement has been finalized, as proposed.
j. Hardcopy Reports for Retests of Hg Low Mass Emission Units
Background
Sections 75.60(b)(6) and (b)(7) require the designated
representative (DR) to submit the results of certain periodic quality-
assurance tests to the appropriate EPA Regional Office or to the State
or local agency, when the test results are requested in writing (or by
electronic mail). In particular, the results of semiannual or annual
RATAs of CEMS and the routine re-tests of Appendix E units may be
requested. If requested, the test results must be submitted within 45
days after the test is completed or within 15 days of the request,
whichever is later. EPA proposed to add a new paragraph (b)(8) to Sec.
75.60, requiring the DR to provide, upon request from EPA or the State,
the results of the semiannual or annual Hg emission tests required
under Sec. 75.81(d)(4) for low-emitting units covered by CAMR. The
proposed time frame for submitting these Hg emission test results would
be the same as the current one for the RATAs and Appendix E re-tests.
Summary of Rule Changes
No adverse comments were received and this provision has been
finalized, as proposed.
k. Wall Effects Adjustment Factors
Background
For sources with flow monitors installed on circular stacks,
reporting of wall effects information is currently required by
Sec. Sec. 75.64(a)(2)(xiii), 75.73(f)(1)(ii)(K) and
75.84(f)(1)(ii)(I), when Method 2H is used in conjunction with Method
2, 2F or 2G. The specific wall effects data elements that must be
reported are found in Sec. 75.59(a)(7)(ii) and (a)(7)(iii). These data
are submitted along with flow RATA results, as supplementary
information.
For rectangular stacks and ducts, some of the same supporting data
elements in Sec. 75.59(a)(7)(ii) and (a)(7)(iii) are needed for flow
RATAs performed using Method 2F or 2G, when wall effects corrections
are applied. Additional supporting data elements, not in the current
rule, are also needed for Method 2 flow RATAs when wall effects
adjustments are made. In view of this, EPA proposed to revise the text
of Sec. Sec. 75.64(a)(2)(xiii), 75.73(f)(1)(ii)(K) and
75.84(f)(1)(ii)(I) and to add RATA support data elements to a new
paragraph, (vii), in Sec. 75.59(a)(7), to clarify which wall effects
data elements must be reported for circular stacks, which ones are
reported for rectangular stacks and ducts, and which data elements must
be reported for both types of stacks.
Summary of Rule Changes
No adverse comments were received on these proposed rule changes
and they have been finalized, as proposed.
F. Subpart H (NOX Mass Emissions)
1. Subpart H Diluent Monitoring Systems
Background
For coal-fired Subpart H units that calculate NOX mass
emissions as the product of NOX concentration and flow rate
and are required to monitor and report the unit heat input, Sec.
75.71(a)(2) requires the installation of an ``O2 or CO2
diluent gas monitor''. Consistent with the definition of a CEMS in
Sec. 72.2, this diluent monitor, which is only used for the heat input
determination, should be described as an ``O2 or CO2
monitoring system''. EPA proposed to revise the text of Sec.
75.71(a)(2) accordingly.
Summary of Rule Changes
No adverse comments were received. This clarification of Sec.
75.71(a)(2) has been finalized, as proposed.
2. Identifying a NOX Mass Methodology
Background
EPA proposed to revise Sec. 75.72 to require that only one
NOX mass emissions methodology be identified in the
monitoring plan at any given time, and to disallow the designation of
primary and secondary NOX mass calculation methodologies.
EPA believes that one methodology for NOX mass emissions is
sufficient. If a source is subject to both Subpart H and to the Acid
Rain Program (ARP) and is concerned about losing NOX data
when the diluent component of the NOX emission rate system
is out-of-control, that source should choose the NOX
concentration times flow rate calculation method as the NOX
mass calculation methodology. This would require a NOX
concentration system to be identified in the monitoring plan, in
addition to the NOX emission rate system. The NOX
concentration system would be used only to determine NOX
mass emissions, and the NOX emission rate system would be
used only to meet the ARP requirement to report NOX in lb/
mmBtu.
Summary of Rule Changes
No adverse comments were received. This provision has been
finalized, as proposed.
3. Reporting of Subpart H Facility Information
Background
Consistent with the proposed revisions to Sec. 75.64, EPA proposed
to revise Sec. 75.73(f)(1), to phase out the requirement of Sec.
75.73(f)(1)(i)(B) to include facility location information in each
quarterly report.
Summary of Rule Changes
No adverse comments were received. This provision has been
finalized, as proposed.
4. Linearity Check Requirements for Ozone Season-Only Reporters
Background
For Subpart H sources that report emissions data on an ozone
season-only (OSO) basis, EPA proposed to revise the linearity check
provisions in Sec. 75.74(c)(2), (c)(2)(i), (c)(2)(ii), (c)(3)(ii),
(c)(3)(vi), and (c)(3)(viii). Historically, OSO reporters have been
required to do a pre-season linearity check, an in-season second
quarter linearity check (in May or June, if the unit operates for >=
168 hours in May and June), and a third quarter linearity check, if the
unit operates for >= 168 hours in that quarter. Many sources have
misunderstood these rule provisions, particularly the requirement to
perform an in-season linearity check in the second quarter. In some
cases, this has resulted in CEMS out-of-control periods and has
required the use of missing data substitution. OSO reporters have also
been required to operate and maintain each CEMS and to perform daily
calibration error tests, in the time period extending from the hour of
completion of the pre-season linearity check through April 30. EPA has
found that this rule provision is also not well-understood by the
affected sources and assessing compliance with the provision has been
difficult, since sources have not been required to report the results
of any off-season calibration error tests done prior to April.
In view of these considerations, EPA proposed to revise Sec.
75.74(c)(2) to require the pre-season linearity checks to be conducted
in the month of April, and to delete all references to
[[Page 4322]]
performing the pre-season linearity checks at other times. The Agency
also proposed to remove the conditional grace period provision from
Sec. 75.74(c)(2)(i)(D), and to address (in Sec. 75.74(c)(3)(ii)(E))
data validation in the case where the April linearity check is not
completed prior to the start of the ozone season. In that case, data
from the monitor would be considered invalid as of May 1, unless the
conditional data validation procedures of Sec. 75.20(b)(3) are
applied. A 168 unit operating hour period of conditional data
validation would be allowed, in which to perform the required linearity
check. Passing the linearity check on the first attempt within the
allotted time would result in the conditionally valid data becoming
quality-assured. Failing the linearity check would result in all data
from the monitor be invalidated back to the beginning of the ozone
season and the data would remain invalid until a linearity check is
passed. Performing the linearity check after the 168-hour period
expires would require the data validation provisions in Sec.
75.20(b)(3)(viii) to be applied, subject to the restrictions of Sec.
75.74(c)(3)(xii).
EPA proposed to add a new paragraph (F) to Sec. 75.74(c)(3)(ii),
stating that a pre-season linearity check done in April fulfills the
second quarter linearity check requirement, and to remove and reserve
related Section 75.74(c)(3)(viii). Further, proposed Sec.
75.74(c)(3)(ii)(B) would require the third quarter linearity check to
be conducted either by July 30 or within a 168 operating hour period of
conditional data validation thereafter. Finally, the Agency proposed
that Sec. 75.74(c)(3)(ii)(G) would address the case where a unit
operates infrequently and the 168 operating hour conditional data
validation period associated with the April linearity check extends
through the second quarter, into the third quarter. In that case, if a
linearity check is performed and passed in the third quarter, before
the 168 operating hour window expires, EPA proposed that this one
linearity check would satisfy all three of the ozone season linearity
check requirements, i.e., for the pre-season, for the second quarter,
and for the third quarter.
Summary of Rule Changes
The amendments to Sec. 75.74(c) have been finalized, as proposed.
Commenters supported EPA's proposal to allow a linearity check
performed in April to satisfy both the pre-season and second quarter
linearity check requirements. However, several commenters requested
that the Agency allow greater flexibility in the timing of the required
linearity checks. The proposed amendments requiring the pre-season
linearity check to be performed April and the 3rd quarter test to be
done in July were perceived as being too restrictive. EPA does not
agree with these commenters that the revised quality assurance
requirements for ozone season-only reporters lack flexibility. The
amendments allow sources to use conditional data validation for up to
168 unit or stack operating hours, in situations where the linearity
check cannot be completed by the prescribed deadline. If the required
test is performed and passed within the allotted window of time, the
source will incur no data loss. OSO reporters desiring greater
flexibility in scheduling quality assurance tests should seriously
consider switching to year-round reporting. Doing so would provide many
benefits, such as grace periods, test deadline extensions, and in some
cases, test exemptions.
5. RATA Requirements for Ozone Season Only Reporters
Background
For Subpart H sources that report NOX mass emission data
on an ozone season-only (OSO) basis, Part 75 has required, for quality-
assurance purposes, that at the start of each ozone season each
required CEMS must be within the ``window'' of data validation of a
current, non-expired RATA. In past years, this requirement has been met
either by performing a RATA in the pre-season (between October 1 and
April 30) or, in some instances, by relying on the results of a RATA
done in the previous ozone season. The rule has further required each
CEMS to be operated, calibrated and maintained in the time period
extending from the completion of the RATA, through April 30. Many
sources choosing the OSO reporting option find this operation and
maintenance (O&M) requirement to be counter intuitive, because they
expect to be required to meet Part 75 monitoring obligations only
during the ozone season.
In view of these considerations, EPA proposed to restrict the
window of time in which pre-season RATAs may be performed. As proposed,
Sec. 75.74(c)(2)(ii) would require the RATAs to be done either in the
first quarter of the year or in the month of April. That restriction
would prohibit RATAs done in the previous year from being used to
validate data in the current ozone season.
EPA also proposed to revise Sec. 75.74(c)(2)(ii)(F), to address
data validation. The proposed data validation rules for RATAs are
similar to those proposed for linearity checks, in that a period of
conditional data validation (720 operating hours) would be allowed when
the pre-season RATA is not completed by the April 30th deadline.
Consistent with these revisions, the Agency proposed to delete the data
validation and conditional grace period provisions in Sec.
75.74(c)(2)(ii)(G) and (c)(2)(ii)(H) and to remove and reserve Sec.
75.74(c)(3)(vi), (vii), and (viii).
Summary of Rule Changes
The amendments to Sec. 75.74(c) have been finalized, as proposed.
One commenter objected to the proposed restriction on the timing of the
RATAs and requested that the existing flexibility in the rule be
retained. The commenter expressed a strong preference to perform RATAs
in the autumn, rather than in the January-April time frame proposed by
EPA. A second commenter stated that EPA should remove the requirement
to keep records of off-season daily calibration and interference check
records in a format suitable for inspection from Sec.
75.74(c)(2)(ii)(E)(1).
Regarding the first commenter's assertion that the proposed RATA
time frame for OSO reporters is too restrictive, EPA recommends that
the owner or operator seriously consider switching to year-round
reporting. Year-round reporting allows complete freedom to schedule
RATAs at any convenient time during the year and provides many
benefits, such as grace periods, test deadline extensions, and in some
cases, test exemptions. Even if EPA had decided not to amend the RATA
provisions for OSO reporters, Sec. 75.74(c)(2)(ii)(E)(1) would still
require the CEMS to be operated, maintained and calibrated in the time
period between the RATA and the start of the next ozone season. Thus,
if the RATAs are performed in the autumn (e.g., November), the CEMS
would have to be maintained and calibrated for at least 10 months of
the year; in this case, OSO reporting offers no clear advantage over
year-round reporting.
EPA did not incorporate the second commenter's suggestion to remove
the recordkeeping requirement from Sec. 75.74(c)(2)(ii)(E)(1).
However, the text of Sec. 75.74(c)(6)(iii) has been revised to remove
the requirement to report the daily calibrations and interference
checks done in the month of April. The requirement to record these data
remains intact, but the reporting has been made optional.
[[Page 4323]]
6. Determining Peaking Status for Ozone Season Only Reporters
Background
EPA proposed to revise Sec. 75.74(c)(11) to clarify that when
peaking unit status for ozone season-only reporters is determined,
3,672 hours (i.e., the number of hours in the ozone season) should be
used instead of 8,760 hours in the capacity factor equation.
Summary of Rule Changes
No adverse comments were received. This provision has been
finalized, as proposed.
7. Calculation of Ozone Season NOX Mass Emissions--LME Units
Background
EPA proposed to correct an organizational error in Subpart H of
Part 75. The proposal would remove Sec. 75.72(f), which describes
ozone season NOX mass calculations for units using the low
mass emission (LME) methodology under Sec. 75.19, and the basic
content of Sec. 75.72(f) would be relocated to Sec. 75.71(e). The LME
provision in Sec. 75.72 appears to have been inadvertently placed in
that section. The monitoring provisions of Sec. 75.72 apply to common
and multiple stack configurations, whereas Sec. 75.71 addresses unit-
level monitoring. LME is a unit-level monitoring methodology.
Summary of Rule Changes
No adverse comments were received. This provision has been
finalized, as proposed.
G. Subpart I (Hg Mass Emissions)
1. Heat Input Provisions for Common and Multiple Stacks
Background
Due to an apparent oversight, the heat input monitoring provisions
for certain monitoring configurations in Subpart I of Part 75 were
inadvertently omitted when Subpart I was promulgated. In particular,
EPA found the heat input methodologies for common stacks shared by
affected and non-affected units and for multiple stack or duct
configurations to be missing. In view of this, the Agency proposed to
add three new paragraphs, (b)(3), (c)(4) and (d)(3) to Sec. 75.82 to
correct this deficiency.
For the common stack shared by affected and non-affected units,
proposed Sec. 75.82(b)(3) would require the owner or operator to
either measure the total heat input rate at the common stack and
apportion it to the individual units by load, according to Sec.
75.16(e)(3), or to determine the heat input rate at the individual
units by installing a flow monitor and a diluent monitor on the duct
leading from each unit to the common stack. For multiple stack
configurations, proposed Sec. 75.82(c)(4) and (d)(3) would require the
owner or operator to determine the hourly unit heat input by measuring
the hourly heat input rate (mmBtu/hr) at each stack, multiplying each
stack heat input rate by the stack operating time (hr) to convert it to
heat input (mmBtu), and then summing the hourly stack heat input
values.
Summary of Rule Changes
No adverse comments were received. These provisions have been
finalized, as proposed.
2. Low Mass Emission Alternative
Background
Section 75.81(b) of Subpart I provides an alternative
(``excepted'') monitoring methodology for units with low Hg mass
emissions. To qualify to use this methodology, emission testing is
required to demonstrate that the unit has the potential to emit no more
than 29 lb (464 ounces) of Hg per year. Once a unit qualifies, periodic
retesting (semiannual or annual, depending on the emission level) is
required to demonstrate that the unit is actually emitting less than 29
lb/yr of Hg.
Section 75.81(e), as originally published, allowed the low mass
emission alternative to be used for common stacks, provided that the
units sharing the stack are tested individually and each one qualifies
as a low-emitter. Though not explicitly stated in the rule, it was
implied that the periodic retests for common stack configurations would
also have to be done at the unit level. EPA has reconsidered this
approach, believing it to be overly restrictive, unnecessarily
difficult, and costly to implement.
Therefore, EPA proposed to revise Sec. 75.81(e) to require Hg
testing of the individual units that share the common stack only for
the initial demonstration that the units individually qualify as low
emitters. Once this has been satisfactorily demonstrated, the required
semiannual or annual retests could then be done at the common stack, at
a normal load level for the configuration.
The proposed revisions to Sec. 75.81(e) would also allow the
initial low mass emitter qualification for a group of identical units
sharing a common stack to be based on emission testing of a subset of
those units. To exercise this proposed option, the group of units would
first have to qualify as identical under Sec. 75.19(c)(1)(iv)(B).
Then, the number of units required to be tested would be determined
from Table LM-4 in Sec. 75.19.
The proposed amendments allowed one exception to the requirement to
test the individual units sharing a common stack, in order to
demonstrate that the units qualify for low mass emitter status, i.e.,
the case where the gas streams from the individual units are combined
together and routed through emission controls that reduce the Hg
concentration (e.g., a wet scrubber) before entering the common stack.
Owners or operators electing to use this option would be required to
perform the testing with all of the units that share the stack in
operation, and the combined load during the testing would have to be
``normal'', as defined in Section 6.5.2.1 of Appendix A.
EPA also proposed to revise Sec. 75.81(c)(1), to specify the
acceptable time frame in which to perform the initial certification
testing for the low mass emission option. As originally published, the
rule simply states that this testing must be done ``prior to the
compliance date in Sec. 75.80(b)'', but does not specify how far in
advance of that date the testing may be done and still be considered
acceptable. Further, Sec. 75.81(d)(1) requires the test results to be
submitted as a certification application, no later than 45 days after
completing the testing. And Sec. 75.81(d)(4) requires periodic Hg
retesting to commence within two or four ``QA operating quarters''
after the quarter of the certification testing.
If there is too long a gap between the certification testing and
the start of the program, it becomes problematic. For instance, if the
testing is done too early, the requirement to submit a certification
application within 45 days could result in applications being submitted
long before the regulatory agencies are ready to receive and process
them. Also, the periodic retesting requirements of Sec. 75.81(d)(4),
which become active on the certification test date, could result in
several Hg retests being done before the program begins. This is
clearly contrary to the purpose of the retests, which, like the
periodic relative accuracy tests of CEMS, are intended to commence
after the compliance date, when Hg emissions reporting has begun. This
also raises questions about which default emission rate to use for the
initial reporting. In view of these considerations, EPA proposed to
revise Sec. 75.81(c)(1), to require that the Hg testing for initial
certification be done no more than 1 year before the compliance date.
Sections 75.81(d)(2) and 75.81(d)(5) would also be revised, to address
the case where a retest may be required before the compliance date
[[Page 4324]]
(e.g., when Sec. 75.81(d)(4) requires a retest within two QA operating
quarters, following a certification test that was done 9 to 12 months
before the compliance date). In such cases, the default Hg emission
rate used at the beginning of the program would be the value that was
obtained in the retest.
Finally, EPA proposed to amend Sec. Sec. 75.81(d)(4) and (d)(5) to
address the emission testing requirements when the fuel supply is
changed. The proposed revisions would require additional Hg retesting
within 720 unit operating hours, following a change in the fuel supply.
The results of this retest would then be applied retrospectively, back
to the time of the fuel switch. The Agency also proposed to revise
Sec. 75.81(c)(1) to require that the fuel combusted during the initial
certification testing be from the same source of supply as the fuel
combusted when the program starts. The proposed revisions only
addressed the emission testing and reporting requirements for one case,
i.e., where the source of supply for the primary fuel (assumed to be
coal) changes. EPA solicited comments and suggestions on how to apply
the Hg low mass emitter option in situations where the coal supply does
not change, but the unit sometimes burns other types of fuel besides
coal or co-fires mixtures of coal and other fuels (i.e., what emission
testing and reporting requirements might be appropriate).
Summary of Rule Changes
Commenters were generally supportive of the proposed amendments
that would reduce the testing requirements for Hg low mass emission
units in common stack configurations. The final rule differs somewhat
from the proposal, however, in that it also allows the initial
qualifying test to be performed at the common stack, if certain
conditions are met. The conditions are: (1) Testing must be done at a
combined load corresponding to the designated normal load level (low,
mid, or high) defined in the monitoring plan; (2) all of the units that
share the stack must be operating in a normal, stable manner and at
typical load levels during the emission testing; (3) the coal combusted
in each unit during the testing must be representative of the coal that
will be combusted in that unit at the start of the Hg mass emission
reduction program (preferably from the same source(s) of supply); and
(4) if flue gas desulfurization and/or add-on Hg emission controls are
used to reduce the level of emissions exiting from the common stack,
these emission controls must be operating normally during the emission
testing and the owner or operator must record parametric data or
SO2 concentration data in accordance with Sec.
75.58(b)(3)(i) to document proper operation of the controls.
For retests, provided that the required load level is attained and
that all of the units sharing the stack are fed from the same on-site
coal supply during normal operation, it is not necessary for all of the
units sharing the stack to be in operation during a retest. However, if
two or more of the units that share the stack are fed from different
on-site coal supplies (e.g., one unit burns low-sulfur coal for
compliance and the other combusts higher-sulfur coal), then the owner
or operator must either: (1) Perform the retest with all units in
normal operation; or (2) if this is not possible, due to circumstances
beyond the control of the owner or operator (e.g., a forced unit
outage), perform the retest with the available units operating and
assess the test results as follows. The Hg concentration obtained in
the retest is used for reporting purposes if the concentration is
greater than or equal to the value obtained in the most recent test.
However, if the retested value is lower than the Hg concentration from
the previous test, then the higher value from the previous test
continues to be used for reporting purposes, and that same higher Hg
concentration is used in Equation 1 to determine the due date for the
next retest.
The final rule expands the testing of groups of identical units
beyond identical units that share a common stack. Section
75.81(c)(1)(iv) has been amended to allow a subset of any group of
identical units to be tested according to Table LM-4 in Sec. 75.19,
whether or not the units share a common stack. This amendment is
modeled after the provisions of Sec. 75.19(c)(1)(iv)(B) for testing
groups of identical LME units.
Several commenters objected to the proposed requirement to perform
retesting of low mass emission units when the fuel supply is changed.
Concerns were expressed that the term ``change in fuel supply'' is not
clearly defined and could be interpreted to require frequent,
unnecessary retesting, especially in light of the variation in coal
supplies from day to day in competitive wholesale power markets. A
number of the commenters recommended that retesting be limited to
changes in coal rank or classification (e.g., changing from bituminous
coal to sub-bituminous coal). EPA has incorporated the commenters'
suggestion into the final rule. Section 75.81(d)(4) of the final rule
clarifies what constitutes a ``change in fuel supply'' that will
trigger LME retesting. If a unit switches to a different rank of coal
as the primary fuel for the unit, in-between the scheduled LME retests
(where coal rank is defined by ASTM D388-99), an additional LME retest
is required within 720 operating hours of the change. The results of
this retest are then applied retrospectively back to the date and hour
of the fuel switch. The four principal coal ranks are anthracitic,
bituminous, subbituminous, and lignitic. The ranks of anthracite coal
refuse (culm) and bituminous coal refuse (gob) are considered to be
anthracitic and bituminous, respectively.
Equation 1 in Sec. 75.81(c )(2), which is used to demonstrate that
a unit qualifies as a Hg low mass emissions unit, conservatively
estimates the unit's potential annual Hg emissions by assuming that it
operates at the maximum potential flow rate for 8,760 hours per year.
One commenter requested that EPA consider modifying Equation 1 to
conditionally allow a number of hours less than 8,760 to be used in the
calculations, the condition being that there is a Federally-enforceable
permit provision in place, limiting the unit's annual operating hours.
EPA has incorporated this suggestion into the final rule. The term
``8,760'' in Equation 1 has been replaced with ``N'', which will either
be 8,760 or the maximum number of operating hours per year allowed by
the unit's Federally-enforceable operating permit (if less than 8,760).
If the operating permit restricts the unit's annual heat input but not
the number of annual unit operating hours, the owner or operator may
divide the allowable annual heat input (mmBtu) by the design rated heat
input capacity of the unit (mmBtu/hr) to determine the value of ``N''.
Finally, no comments were received on the proposal to require that
the Hg emission testing for initial certification of a low mass
emission unit be done no more than 1 year prior to the applicable
compliance date. Therefore, this provision has been finalized, as
proposed. For units subject to the Clean Air Mercury Regulation (CAMR),
the certification deadline is January 1, 2009. In view of this, only
those Hg emission tests of candidate low mass emission units that are
performed on and after January 1, 2008 will be accepted for initial
certification.
3. Harmonization of Subpart I With Other Proposed Rule Revisions
Background
Subpart I of Part 75 also contains a recordkeeping and reporting
section (Sec. 75.84). which, for the most part, cross-references the
primary monitoring
[[Page 4325]]
plan, recordkeeping, notification and reporting sections of the rule
(i.e., Sec. Sec. 75.53, 75.57 through 75.59, 75.61, and 75.64) and
other sections of Subpart I.
To make Subpart I consistent with the proposed revisions to the
monitoring plan, recordkeeping, notification, and reporting sections of
Part 75, EPA proposed to make a number of minor adjustments to the text
of Sec. Sec. 75.84(c)(3), (e)(1), (e)(2), and (f)(1).
Summary of Rule Changes
No adverse comments were received. These provisions have been
finalized, as proposed.
H. Appendix A
1. CO2 Span Values
Background
EPA proposed to revise Section 2.1.3 of Appendix A, to allow the
use of CO2 spans less than 6.0 percent CO2 if a
technical justification is provided in the hardcopy monitoring plan.
This added flexibility in the CO2 span value mirrors a
similar provision in Section 2.1.3 for O2 span values.
Summary of Rule Changes
No adverse comments were received. This provision has been
finalized, as proposed.
2. Protocol Gas Audit Program
Background
EPA is responsible for implementing air quality programs that rely
heavily on the accuracy of calibration gas standards. Section 2.1.10 of
``EPA Traceability Protocol for Assay and Certification of Gaseous
Calibration Standards'' (Protocol Procedures), September 1997 (EPA-600/
R-97/121) states that EPA will periodically assess the accuracy of
calibration gases and publish the results. Between 1978 and 1996, EPA
conducted several performance audits of calibration gases from various
manufacturers. One notable result of these audits was a steady,
significant reduction in the failure rate of the audited gas cylinders,
from about 27% in 1992 down to 5% in 1996. The annual audits were
discontinued after 1996. Then, in 2003, EPA conducted a ``surprise''
audit of 14 national specialty gas producers and found that the failure
rate had risen to 11%.
In view of this, EPA proposed to establish a Protocol Gas
Verification Program (PGVP) and would require that EPA Protocol Gases
being used for 40 CFR Part 75 purposes be obtained from specialty gas
producers who participate in the PGVP. As proposed, the rule would
allow only program participants to market their gas standards as ``EPA
Protocol Gases.'' EPA proposed to maintain a web site, listing the PGVP
participants and the audit results, in order to provide calibration gas
users with detailed information about the quality of EPA Protocol
Gases.
EPA also proposed to: (1) Add a definition of ``specialty gas
producer'' to Sec. 72.2; (2) delete several calibration gas standards
and reference materials from section 5.1 of appendix A (believing them
to be prohibitively expensive and not used in practice by Part 75
sources); (3) remove from Sec. 72.2 the corresponding definitions of
the deleted calibration gas standards; and (4) consolidate the
remaining calibration gas standards under section 5.1 of appendix A.
Finally, EPA requested comment on the appropriate accuracy
specification to apply to Hg cylinder gases and other Hg calibration
standards (e.g., gases from NIST-traceable generators). Currently, EPA
requires that accuracy of other EPA Protocol gases to be within 2
percent of the certified tag values.
Summary of Rule Changes
Only one organization commented on the proposed protocol gas
verification program (PGVP). The commenter stated that a transition
period is needed to implement the program. Sources need time to
communicate with their gas vendors regarding their participation in the
PGVP. The commenter further asserted that the PGVP would be disruptive
and costly, both in the short-term and in the long-term, and that the
affected sources would bear the brunt of the cost impact.
EPA agrees with the commenter regarding the need for a transition
period. The final rule amends section 5.1.4 (c) to have the Protocol
Gas Verification Program (PGVP) take effect on January 1, 2009. As the
commenter has stated, the costs of the PGVP will be borne by the Part
75 sources using the calibration gases, and the Agency notes that these
minimal costs ($5 to $10 added to a $500 to $1,000 cylinder) will be
offset by the savings generated by fewer failed calibration error
tests, linearity checks, and relative accuracy test audits.
3. Requirements for Air Emission Testing Bodies
Background
Since the inception of the Acid Rain Program, field audits of Part
75-affected facilities have brought to EPA's attention a number of
improperly-performed RATAs and other QA/QC tests. In view of this, EPA
proposed to revise Section 6.1 of Appendix A to require all individuals
who perform the emission tests and CEMS performance evaluations
required by Part 75 to demonstrate conformance with ASTM D7036-04
``Standard Practice for Competence of Air Emission Testing Bodies''.
ASTM D7036-04 specifies the general requirements for demonstrating that
an air emission testing body (AETB) is competent to perform emission
tests of stationary sources.
Proposed revisions to Section 6.1.2 of Appendix A, Section 2.1 of
Appendix E, and Section 1 of Appendix B make it clear that this
requirement would apply only to AETBs that perform RATAs,
NOX emission tests of Appendix E and LME units, or Hg
emission tests of low-emitting units. It would not be applicable to the
daily operation, daily QA/QC (daily calibration error check, daily flow
interference check, etc.), weekly QA/QC (i.e., Hg system integrity
checks), quarterly QA/QC (linearity checks, etc.), and routine
maintenance of the CEMS.
EPA also proposed to incorporate ASTM Method D7036-04 by reference
in Sec. 75.6(a)(45), and to add a definition of ``Air Emission Testing
Body'' to Sec. 72.2.
Summary of Rule Changes
The amendments to Section 6.1.2 of Appendix A, Section 2.1 of
Appendix E, and to Section 1 of Appendix B, requiring AETBs to conform
to ASTM D7036-04, have been finalized, as proposed. Two commenters
strongly supported the proposed revisions. However, several others
objected to them, believing they would be costly and burdensome,
without producing any noticeable improvement in data quality. EPA does
not agree with these commenters, for the following reasons.
The experience of the State and Federal regulators in the ASTM work
group indicates that implementation of the ASTM Practice will result in
improved data quality. EPA believes the evidence is abundant that
unqualified, under-trained and inexperienced testers are often deployed
on testing projects. The Agency has had experiences with tests that
have been invalidated or called into question due to poor performance
by testing contractors (see Docket Items OAR-2005-0132-0009, -0021, and
-0035). Conformance with ASTM D7036-04 does not guarantee that every
test will be performed properly. However, it will reduce the likelihood
of problems. Furthermore, it provides a guideline for both regulatory
agencies and affected sources to evaluate and select competent testing
[[Page 4326]]
firms. One of the cornerstones of the Practice is that AETBs must
collect performance data on how well they plan and execute test
projects. These data must be shared with regulators and clients upon
request.
In response to claims that ASTM D7036-04 will significantly
increase the cost and burden of Part 75 testing, EPA notes that no data
were provided to support these claims. The ISO 17025 standard upon
which the ASTM standard is based has been implemented in Europe for
many years. Mark Elliot, Chairman of the Stack Testing Association
(STA) of Great Britain, has provided the following information on the
costs of their programs. Their certification program (for individuals)
is called MCERTS.
MCERTS testing fees: Level 1 $350; Level 2 $940
Technical endorsements (1-4): $350 each
The Level 2 certification requires a personal interview with the
applicant. Please note that according to Mr. Elliot, this program has
been successfully implemented in the UK with no small companies going
out of business and no complaints of being overly burdensome on
industry. In fact, many large companies such as Mobil, Dow, Pfizer, and
3M are members of the STA and fully support the program because,
according to Mr. Elliot, they believe it improves the quality of the
data provided by testing companies. Even major UK utility companies
such as Drax Power, Energy Power Resources, the Electricity Supply
Board, PB Power, Scottish and Southern Energy, and Scottish Power
participate in the program. And they do this voluntarily because they
have found it to their benefit to do so.
There are several differences between the program described in the
final rule and the UK program. First, the final rule does not require
accreditation. The individual testing requirements in the rule are less
expensive and less stringent than the UK program. In the US, The Source
Evaluation Society is currently providing Qualified Individual testing.
The fees are $155 for the first test (including a one-time $15 SES
membership) and $89 for any subsequent tests taken during the same
testing session). It should also be noted that ASTM D7036-04 does not
require that every individual be tested. Only one ``Qualified
Individual'' need be present on-site during a test. Therefore, even
this minimal cost and burden is considerably less than the successful
UK program.
The costs of coming into initial compliance with the ASTM D7036-04
standard depend on the current state of an AETB's quality program.
Those that do not currently have an organized quality program will most
likely incur greater costs than those who do. In any case, the burden
will be no greater than that experienced by the UK companies who
successfully went through the same process.
The main costs to comply with the ASTM D7036-04 standard are
associated with taking a stack test QSTI (qualified stack test
individual) competency exam, and developing or revising a quality
assurance (QA) manual. A nationwide compliance cost estimate may be
obtained using the following estimates:
450 stack test companies in U.S. (The number of private
(external) stack test companies came from http://www.epa.gov/ttn/emc/software.html#testfirm.
RMB Consulting, Inc. estimated 10 in-house
utility RATA test teams in the U.S.);
On average, 10 people per company (Source: http://www.epa.gov/ttn/emc/software.html#testfirm
);
QSTI exam (required by ASTM) costs $150 and must be taken
every 5 years (Source: December 11, 2006 letter from the Source
Evaluation Society in Docket OAR-2005-0132); and
Roughly 1 QSTI is required for every 3 people in a stack
test company.
Using these inputs, the Agency estimates the cost to comply with
ASTM D7036-04 at about $100 per yr per company to cover the QSTI exam.
There is also approximately a $4,000 one time cost per company, whether
a large or small entity as defined by the Small Business
Administration's (SBA) regulations at 13 CFR 121.201, to develop a QA
manual (estimate provided by Air Tech, see Docket Item EPA-
HQ-OAR-2005-0132-0093). However, the costs will be borne by the Part 75
sources using the air emission testing bodies, and the Agency notes
that these costs will be offset by the savings generated by fewer
failed or incorrectly performed relative accuracy test audits, and
fewer repeat tests required. Therefore, the effect of this revision is
to actually relieve a regulatory burden on these entities.
Regarding the issue of the financial impact on smaller companies
and the request to provide funds to these companies, EPA notes that
small stack test companies were represented on the ASTM work group. At
least one small stack test company (3 people) has already complied with
ASTM D7036-04, is supportive of the requirement, and expects to
actually realize an increase in business because of their compliance
with ASTM D7036-04. As stated in another response, the costs to comply
with ASTM D7036-04 are reasonable. Similar requirements have been
successfully implemented for many years in the UK with no small
companies going out of business and no complaints of being overly
burdensome on industry. EPA does not expect to provide funds to support
small stack test companies in meeting the requirements of ASTM D7036-
04.
EPA notes that virtually the same program has been in place in
Europe for several years and is functioning very well with the support
of stack testers, the government, and industry. The ASTM standard is
actually less stringent in some areas than the European program. Based
on this extensive experience in Europe, EPA believes that this program
can be successfully implemented here in the U.S. with very little
additional burden. In summary, there is an abundance of both data and
experience showing that this program can be implemented without an
unreasonable burden, and also (according to UK industry participants)
that it will improve the quality of data.
Two commenters asserted that the existing infrastructure is not
adequate for testers to comply with the ASTM method. EPA disagrees with
these claims. The Source Evaluation Society is currently offering
qualification exams in several areas. The commenters may be concerned
that the SES website used to state that their exams may not
specifically satisfy the requirements of the ASTM Practice (because
they were not developed specifically for that purpose). However, SES
has updated the wording on their Web site to say that their
qualification exams do meet the exam requirement of the ASTM Practice.
The Stack Testing Accreditation Council (STAC) also recognizes that not
only does the SES program meet the requirements of the ASTM standard--
it actually exceeds them. It requires more experience than the ASTM
standard and also requires letters of recommendation. Both EPA and STAC
accept an SES certification as meeting the external testing and
experience requirements of the ASTM Practice.
If an external QSTI test is not available to a company, an internal
test may be used to meet the requirements of ASTM D7036-04 until an
external test becomes available. EPA is aware of at least one large
stack test company that has developed a training module for mercury
methods meeting the requirements of the ASTM D7036-04, and has trained
and tested their people according to the internal qualification exam
provision of ASTM D7036-04. When a third party test becomes
[[Page 4327]]
available, this company has indicated that they will re-certify their
people according to the requirements of ASTM D7036-04. The Source
Evaluation Society is reviewing steps to improve and expand the QSTI
examination process.
Four commenters asked EPA to clarify how compliance with ASTM
D7036-04 would be determined. Section 6.1.2 in Appendix A of the final
rule specifically states that there are two ways an AETB can certify
compliance: (1) A certificate of accreditation, or (2) a letter of
certification signed by senior management. The latter option is similar
to the way major sources certify compliance with their Title V permits.
However, AETBs are under much more direct regulatory scrutiny than a
Title V source. Every state has a field test observer program. In the
case of one large stack testing company, Clean Air Engineering, about
half of their compliance tests are directly observed by state
regulators. This oversight provides an on-going check of whether an
AETB remains in conformance. In co-operation with the New Jersey DEP, a
standardized state observer checklist is being developed that will
facilitate incorporating state observer assessments into the ASTM
process.
EPA expects to treat non-compliance with this standard in the same
way it treats noncompliance with any other standard--using its
enforcement discretion. EPA does not anticipate invalidating test
results because of minor infractions. The proper way to deal with these
issues, if either the regulatory authority or the client discovers
them, is to notify the AETB that a problem has been found. The AETB is
then obligated to initiate a corrective action to address the problem.
This becomes part of the AETB's Performance Data required by the
Practice. The Agency recommends that the client also ask the AETB to
report back on what corrective actions were taken. In the case of
serious infractions, EPA may exercise the same authority it has always
had to reject the test.
EPA encounters deviations in test methodology routinely in
reviewing stack test reports. Minor deviations are noted and reported
back to the source but the underlying results are accepted. Major
deviations result in a rejection of the test. This situation is no
different. This Practice should be treated much like a test method in
this regard. Minor deviations may be of the type the commenters cite in
their examples. Major deviations may include (for example) not having a
Qualified Individual on-site, not having proper calibration records for
the equipment used, or failing to follow through with corrective
actions when required.
There will undoubtedly be some discussions between EPA, affected
sources and AETB's as this program unfolds that will help define the
implementation of the Practice. But this is the case with every new
rule and standard.
There is always a balance in standard writing between being overly
detailed and prescriptive and being too loose and flexible. The
stakeholders involved in the consensus process of ASTM determined that
the proper balance had been achieved. It is important to keep in mind
that ASTM D7036-04 is essentially an international standard that has
been used successfully in countries all over the world.
Three commenters requested that EPA provide a 1-2 year transition
period after promulgation of the final rule, to allow AETBs sufficient
time to conform to ASTM D7036-04. Particular concerns were expressed
about the availability of Qualified Individuals (QIs) for Hg emission
testing. EPA agrees that a transition period is appropriate, given the
testers' relative unfamiliarity with Hg test methods. Therefore, the
final rule gives AETBs until January 1, 2009 to comply with ASTM D7036-
04.
A number of other comments were received on the proposed AETB
certification program. These are addressed in detail in the Response to
Comments (RTC) document.
4. Linearity Requirements for Dual-Span Applications
Background
In May 1999, EPA revised the linearity check provisions in Part 75,
Appendix A, section 6.2, to exempt SO2 and NOX
span values of 30 ppm or less from performing linearity checks. Since
the May 1999 revisions became effective, some have questioned whether
the linearity exemption applies only to ongoing QA or whether it
applies also to initial certification. Others have asked whether the
exemption applies only to a particular measurement range or to all of
the linearity check requirements for a monitoring system. In view of
this, EPA proposed to revise Section 6.2 of Appendix A to make it clear
that the 30 ppm linearity exemption: (1) Is range-specific; (2) covers
both initial certification and ongoing QA; (3) does not remove the
requirement to perform linearity checks of the high range (if > 30 ppm)
for dual span applications; and (4) does not take away the linearity
check requirements for the diluent monitor component of a
NOX-diluent monitoring system.
Summary of Rule Changes
The proposed amendments to Section 6.2 of Appendix A have been
finalized, without substantive change. At the request of one commenter,
the final rule clarifies that the low-span linearity exemption applies
to recertification as well as to initial certification and ongoing QA.
5. Dual Span Applications-Data Validation
Background
EPA proposed to clarify the relationship between the quality-
assured (QA) status of the low and high ranges of a gas monitor in a
dual-span application. Sections 2.1.1.5(b) and 2.1.2.5(b) of Appendix A
have provided instructions for reporting SO2 and
NOX concentration data when the full-scale range of the
monitor is exceeded. For single-range applications, reporting a value
of 200 percent of the range has been required when a full-scale
exceedance occurs. For dual range applications, if the low range is
exceeded, no special reporting has been necessary, provided that the
high range is ``available and not out-of-control or out-of-service for
any reason''. However, if the high range is ``not able to provide
quality-assured data'' during the low-range exceedance, then sources
have been required to report the maximum potential concentration (MPC).
Believing that the two phrases used to describe the QA status of
the high range during low-scale exceedances, i.e., ``available and not
out-of-control or out-of-service for any reason'' and ``not able to
provide quality assured data'' to be too general, the Agency proposed
to revise these rule texts by defining the QA status of the high range
in terms of its most recent calibration error and linearity checks.
Provided that both of these QA tests are still ``active'', i.e., their
windows of data validation have not expired, the high range would be
considered in-control and able to provide quality-assured data. However
if either of the tests has expired, data recorded on the high range
would be considered invalid until the expired test was repeated and
passed. The MPC would be reported until the expired high-range test is
redone or until the data return to the low scale. Thus, the proposed
revisions would clarify that when the low range is up-to-date on its QA
tests but the high range is not, the QA status of each range is
evaluated separately.
[[Page 4328]]
Summary of Rule Changes
No adverse comments were received. These provisions have been
finalized, as proposed.
6. Cycle Time Test-Stability Criteria
Background
The cycle time test described in Section 6.4 of Appendix A is
required for the initial certification and recertification of gas
monitoring systems, and occasionally as a diagnostic test. The test is
designed to determine how long it takes for a monitor to respond to
step changes in gas concentration. Two calibration gases (zero- and
high-level) are used for the test, which has both an upscale and a
downscale component.
Section 6.4 has specified criteria for determining when a stable
gas concentration reading has been obtained. The reading is considered
stable if it changes by less than 2.0 percent of the span value for 2
minutes or less than 6.0 percent from the average concentration over 6
minutes. These criteria are reasonable when the source effluent
concentrations are moderate or high. However, when concentrations are
very low, the criteria can become overly stringent and difficult to
meet. In view of this, the Agency proposed to add alternative stability
criteria to Section 6.4 of Appendix A. By the alternative criteria, an
SO2 or NOX reading would be considered stable if
it changed by no more than 0.5 ppm for 2 minutes or, for a diluent
monitor, if it changed by no more than 0.2% CO2 or
O2 for 2 minutes.
Summary of Rule Changes
Substantive changes have been made to the cycle time test
procedure, in response to comments received. The sequence of the test
has been reversed, i.e., it now begins with a stable reading of stack
emissions and ends with a stable reading of calibration gas
concentration (see section 2.6 of the Response to Comments document for
further discussion). Commenters were generally supportive of the
proposed alternative stability criteria, and these have been
incorporated into the final rule. One commenter noted the absence of
corresponding alternative stability criteria for Hg monitors. To
correct this apparent oversight, the final rule includes an alternative
specification of 0.5 [mu]g/m\3\ for Hg CEMS. The same commenter also
expressed concerns about temporal variations in stack gas concentration
(particularly for Hg) that can make it difficult to meet the stability
criteria, and recommended that the order of the cycle time test be
reversed, i.e., begin the test by measuring stack gas emissions and
then inject the calibration gas. EPA agrees with this comment and has
revised the cycle time test procedure and Figure 6 in Appendix A
accordingly. EPA believes this change in the test procedure (which is
closer to the way in which the test was originally presented in the
January 1993 rule) gives a more accurate indication of the monitor's
true response time and will help to prevent ``false positive'' test
failures.
EPA has also revised the reporting requirement (in Appendix A Sec.
6.4) for cycle time tests of dual range monitors in light of the
transition to the revised XML format. The change requires that cycle
time for both ranges of a component be reported separately (consistent
with the reporting of other component level tests for CEMS), rather
than only reporting the results from the range with the longer cycle
time. This change is consistent with the proposed changes that required
reporting of certain test at the component level rather than at a
system/component level, which overall reduces redundant reporting of
test data from shared components. No adverse comments were received on
those similar proposed changes. This revision was necessary for
consistency with those other proposed changes which EPA is finalizing.
7. System Integrity and Linearity Checks of Hg CEMS
Background
The required certification tests for a Hg CEMS include a 3-level
system integrity check, using a NIST-traceable source of oxidized Hg
and a 3-level linearity check, using elemental Hg standards. The
performance specification for the system integrity check, which is
found in paragraph (3)(iii) of Appendix A, Section 3.2, has been that
the system measurement error must not exceed 5.0 percent of the span
value at any of the three calibration gas levels. However no
explanation of how to calculate the measurement error has been
provided. EPA proposed to restructure paragraph (3) of Section 3.2, to
add the necessary mathematical procedure.
Believing that the performance specification for the linearity
check (which is done with elemental Hg) should be at least as stringent
as the performance for the system integrity check (which is done with
oxidized Hg), the Agency also proposed to make the linearity and system
integrity check specifications for Hg monitors the same, i.e., 5.0
percent of the span value, with an alternative specification to 0.6
[mu]g/m\3\ absolute difference between the reference gas value and the
monitor response.
Summary of Rule Changes
In the final rule, the performance specifications for the linearity
checks and system integrity checks of Hg monitors have been made the
same, but the proposed 5.0 percent of span criterion (with an
alternative specification of 0.6 [mu]g/m\3\) has not been adopted. The
commenters did not take issue with the proposal to equalize the
performance specifications for the two QA tests, but several commenters
objected to the proposed values of the specifications, citing a lack of
supporting data to demonstrate that the specifications are achievable.
Two commenters favored setting both specifications at the existing
values for the linearity check, i.e., 10.0 percent of the reference gas
value, with an alternative specification of 1.0 [mu]g/m\3\.
In response to these comments, EPA analyzed data from two recent
field studies in which elemental and oxidized Hg calibration gases were
injected into commercially-available Hg CEMS, at different
concentration levels (low, mid, high). Based on the results of the data
analysis, the Agency has concluded that equalizing the performance
specifications for linearity checks and system integrity checks of Hg
monitors at 10.0 percent of the reference gas value, with an alternate
specification of 0.8 [mu]g/m\3\ absolute difference is appropriate, and
the final rule incorporates these specifications.
A total of 97 data points from the two field studies were analyzed.
Data recorded during known periods of probe malfunction and excessive
analyzer drift were excluded from the analysis. Eighteen of the 97 data
points analyzed were elemental Hg injections, and the rest were
oxidized Hg injections. Each gas injection was evaluated on a pass/fail
basis against six candidate sets of performance specifications. These
were: (1) The proposed performance specifications, i.e., 5.0 percent of
span, with an alternative specification of 0.6 [mu]g/m\3\; (2) the
existing linearity specifications, i.e., 10.0 percent of the reference
gas value, with alternative specification of 1.0 [mu]/m\3\; (3) the
existing system integrity specification, i.e., 5.0 percent of span,
with no alternative specification; (4) 5.0 percent of span, with an
alternative specification of 0.8 [mu]g/m\3\ ; (5) 5.0 percent of span,
with an alternative specification of 1.0 [mu]g/m\3\; and (6) 10.0
percent of the reference gas value, with alternative specification of
0.8 [mu]g/m\3\. For each set of performance specifications, the pass
rate of the 97 gas
[[Page 4329]]
injections was determined. The two highest pass rates (96.9% and 95.9%)
were attained with sets (2) and (5), respectively, which have the
widest alternative specification of 1.0 [mu]g/m\3\. Similarly high pass
rates (93.8% and 94.8%) were also attained with sets (4) and (6), both
of which have an alternative specification of 0.8 [mu]g/m\3\. The
lowest pass rates (85.5% and 75.3%) were attained with sets (1) and
(3), the proposed performance specifications and the existing system
integrity check specification.
From these results, EPA concludes, on the one hand, that both the
proposed performance specifications (set 1) and existing system
integrity check specifications (set 3) may be too stringent. On the
other hand, very high pass rates were achieved with the four sets
having the wider alternate specifications of 1.0 [mu]g/m\3\ and 0.8
[mu]g/m\3\, i.e., sets (2), (5), (4), and (6). For these four sets, it
seems to make little or no difference whether the main specification is
5.0 percent of span or 10.0 percent of the reference gas value. In view
of these considerations, EPA has selected the main specification for
the system integrity and linearity checks to be 10.0 percent of the
reference gas value, and the alternative spec