United States '
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-453/R-99-004b<
May 1999
Air
& EPA
NATIONAL EMISSION STANDARDS FOR
HAZARDOUS AIR POLLUTANTS FOR
SOURCE CATEGORIES : OIL AND NATURAL
GAS PRODUCTION AND NATURAL GAS
TRANSMISSION AND STORAGE -
BACKGROUND INFORMATION FOR FINAL
STANDARDS
SUMMARY OF PUBLIC COMMENTS AND
RESPONSES
-------
EPA-453/R-99-004b
National Emission Standards for Hazardous Air Pollutants for Source Categories: Oil and
Natural Gas Production and Natural Gas Transmission and Storage
Background Information for
Promulgated Standards - Summary of
Public Comments and Responses
Emission Standards Division
U.S. Environmental Protection Agency
Region 5, Liorary (PL-12J)
77 West Jackson Boulevard, 12th Ftoof
Chicago, IL 60604-3590
U. S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
-------
May 1999
Disclaimer
This report is issued by the Office of Air Quality Planning and Standards, U. S.
Environmental Protection Agency. Mention of trade names and/or commercial products is not
intended to constitute endorsement or recommendation for use. Copies of this report are
available free of charge to Federal employees, current contractors and grantees, and nonprofit
organizations-as supplies permit-from the Library Services Office (MD-35), U. S.
Environmental Protection Agency, Research Triangle Park, NC 27711, (919-541-2777) or, for a
nominal fee, from the National Technical Information Service, 5285 Port Royal Road,
Springfield, Virginia 22161, (703-487-4650).
111
-------
IV
-------
ENVIRONMENTAL PROTECTION AGENCY
National Emission Standards for Hazardous Air Pollutants for Source Categories: Oil and
Natural Gas Production and Natural Gas Transmission and Storage
Background Information for Promulgated Standards -
Summary of Public Comments and Responses
1 The final National Emission Standards for Hazardous Air Pollutants (NESHAP) will regulate
emissions of hazardous air pollutants from oil and natural gas production and natural gas
transmission and storage . Only those operations that are part of major sources under
section 112(d) of the Clean Air Act as amended in 1990 will be regulated.
2 Copies of this document have been sent to the following Federal Departments: Labor, health
and Human Services, Defense, Transportation, Agriculture, Commerce, interior, and Energy;
the national Science Foundation; and the Council on environmental Quality; members of the
State and Territorial Air Pollution program Administrators; the Association of Local Air
Pollution Control Officials; EPA Regional Administrators; and other interested parties.
3 For additional information contact:
Mr. Greg Nizich
Waste and Chemical Processes Group (MD-13)
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919)541-3078
4. Paper copies of this document may be obtained from:
National Technical Information Service (NTIS)
5285 Port Royal Road
Springfield, VA 22161
Telephone: (703) 487-4650
U. S. EPA Library Services Office (MD-35)
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
5. Electronic copies of this document may be obtained from the EPA's OAR Technology
Transfer Network website (TTNWeb).
The TTNWeb is a collection of related Web sites containing information about many areas of
air pollution science, technology, regulation, measurement, and prevention. The TTNWeb is
directly accessible from the Internet via the World Wide Web at the following address:
http//www.epa.gov/ttn
-------
TABLE OF CONTENTS
Page
1.0 SUMMARY 1-1
1.1 BACKGROUND 1-1
1.2 SUMMARY OF SIGNIFICANT CHANGES SINCE PROPOSAL . . . 1-1
1.2.1 Area Source Regulation j.-i
1.2.2 Definition of Facility l -2
1.2.3 Potential-to-Emit 1-3
1.2.4 Averaging Periods 1- 6
1.2.5 Process Modifications 1-7
1.2.6 Standards for Natural Gas Transmission and
Storage 1-8
1.2.7 Monitoring. Recordkeeping. and Reporting
Requirements 1-11
1.3 SUMMARY OF IMPACTS OF PROMULGATED ACTION .... 1 -12
2.0 SUMMARY OF PUBLIC COMMENTS
2.1 APPLICABILITY
2.1.1 Determination of Major Source Status
2.1.2 Exemptions
2.1.3 Other Applicability Issues ....
2.2 DEFINITIONS
2.2.1 Facility
2.2.2 Other Comments on Definitions ± 56
2.3 ASSOCIATED EQUIPMENT
2.4 HAP EMISSION POINTS
2.5 IMPACTS
2.5.1 Cost Impacts Including Production Recovery
Credits 2 100
2.5.2 Environmental Impacts 2-107
2.6 ECONOMIC ANALYSIS 2-109
2.7 LEGAL ISSUES [OTHER THAN ISSUES ASSOCIATED WITH THE
EPA'S INTERPRETATION OF SECTION 112(n)(4)(A) AND (B)]
2-114
2.8 PERMIT ISSUES 2-116
2.9 ENFORCEMENT ISSUES 2-120
2.10 CONTROLS 2 -132
2.10.1 MACT Floor 2-132
2.10.2 Averaging Period 2-140
2.10.3 Process Vent Standards 2-146
2.10.4 Equipment Leak Standards 2-152
2.10.5 Control Device Requirements 2-155
2.10.6 Storage Vessel Standards 2-157
2.11 MONITORING, RECORDKEEPING, AND REPORTING .... 2-159
vii
-------
TABLE OF CONTENTS
Page
2.11.1 Monitoring Requirements 2 -161
2.11.2 Recordkeeping and Reporting Requirements
2-181
2.12 TEST METHODS 2-194
2.13 COMPLIANCE 2-201
2 .13.1 Compliance Procedures 2-201
2.13.2 Compliance Determination 2-205
2.13.3 Compliance Dates 2- 219
2.14 WORDING OF REGULATIONS (OTHER THAN APPLICABILITY AND
DEFINITIONS) 2-221
2.15 GENERAL PROVISIONS 2 -226
2.16 MISCELLANEOUS 2-232
2.16.1 Health Effects 2 -232
2.16.2 Other Miscellaneous Comments 2-237
2.17 GENERAL COMMENTS SPECIFIC TO SUBPART HHH (NOT OTHERWISE
ADDRESSED) 2-240
2.18 COMMENTS RECEIVED ON THE JANUARY 15, 1999 SUPPLEMENTAL
NOTICE (64 FR 2611) 2- 254
vill
-------
1.0 SUMMARY
1.1 BACKGROUND
On February 6, 1998, the Environmental Protection Agency
(EPA) proposed standards of performance for the oil and natural
gas production source category (63 FR 6288) under authority of
Section 111 of the Clean Air Act.
Public comments were requested on the proposal in the
Federal Register. There were 54 commenters composed mainly of
industry and trade associations. Also commenting were State and
local agencies, consultants and engineers, environmental groups,
and other interested parties.
The written comments were submitted, along with the
responses to these comments, are summarized in this document.
The summary of comments and responses serves as the basis for the
revisions made to the standard between proposal and promulgation,
1.2 SUMMARY OF SIGNIFICANT CHANGES SINCE PROPOSAL
In response to comments received on the proposed standards,
several changes have been made to the final rules. A summary of
the substantive changes made since the proposal in response to
comments is provided in the following sections. Additional
information on the final rules is contained in the docket for
this rule (Air Docket A-94-04).
1.2.1 Area Source Regulation
In the February 6, 1998 Federal Register notice (63 FR
6291), the EPA gave notice of its intention to add oil and
natural gas production as an area source category, but did not
amend the source category list to include such a category.
In order to ensure that regulations applicable to the area
source category are consistent with the Urban Air Toxics
Strategy, to be implemented under section 112(k) of the Act, the
EPA has deferred the regulation of oil and natural gas production
facilities which are area sources until the Urban Air Toxics
Strategy is finalized. The EPA expects this strategy to be
finalized later this year.
1-1
-------
Several comment letters were received regarding the area
source regulation. Since the regulation of area sources has been
deferred, summaries of these comment letters and the EPA's
responses to these comments are not included in this document.
1.2.2 Definition of Facility
The EPA developed the proposed definition of facility to (1)
identify criteria that define a grouping of emission points that
meet the intent of the language contained in section 112(a)(1) of
the Act: "... located within a contiguous area and under
common control, . . ."; and (2) contain terms that are meaningful
and easily understood within the regulated industries. The
proposed definition was based on individual surface sites and the
idea that equipment located on different oil and gas properties
(oil and gas lease, mineral fee tract, subsurface unit area,
surface fee tract, or surface lease tract) shall not be
aggregated. In addition, the proposed definition of a production
field facility was limited to glycol dehydration units and
storage vessels with the potential for flash emissions.
Several commenters responded to the EPA's request for
comments on the definition of facility. The commenters requested
clarification of, or suggested changes to, the proposed
definition of facility.
In response to comments regarding specific clarification to
the definition of facility, the EPA has made several changes to
the definition of facility. The EPA modified the definition of
facility to point to the definition of "surface site." In
subpart HHH, the EPA has added a definition of "surface site,"
and modified the definition of facility to point to the new
definition of "surface site."
The EPA further modified the definition of facility in
subpart HH by: (1) specifying that "upgraded91 means "the removal
of impurities or other constituents to meet contract
specifications"; (2) changing the term "unit areas" to "surface
unit areas"; and (3) specifying that separate surface sites,
whether or not connected by a road, waterway, power line or
pipeline, would not be considered a part of the same facility.
1-2
-------
Other commenters requested that the EPA clarify, within the
definition of facility in subpart HHH, whether the EPA intended
to exclude facilities used to store natural gas after the gas
enters the local distribution system of a gas utility. The
commenter recommended that the EPA clarify that the definition of
facility applies all the way to the end user only if there is no
local distribution company.
The affected source in the natural gas transmission and
storage source category should run all the way to the end user
only if there is no local distribution company. Therefore, the
EPA modified the definition of facility in subpart HHH to state
that if there is not a local distribution company, the facility
runs to the end user.
1.2.3 Potential-to-Emit
Several commenters were concerned with the methods used to
determine whether or not a facility was a major source. In
particular, the EPA received several comment letters regarding
the calculation of a facility's potential-to-emit (PTE) when
determining a facility's major source status. The EPA received
comments stating that the calculation of PTE should not be based
on equipment operating capacity because it would result in
overregulation, but should consider the inherent operating
limitations of the facility (e.g., declining production levels
over time). Other commenters recommended that the EPA should
provide a simplified approach to calculate PTE, which takes into
account design and operational limitations.
Several commenters were concerned that PTE estimates, as
defined in the General Provisions, would be unrealistically high
and would subject many small,insignificant sources to the NESHAP
requirements. The commenters requested that PTE be based on the
inherent design and operational limitations of production and
transmission and storage facilities, such as throughput rates.
According to commenters, the throughput of oil and natural
gas production operations declines over time, and existing
equipment is often designed, constructed and operated based on
high initial production rates. Therefore, the commenters
1-3
-------
suggested that the facilities are usually operated at actual
throughput rates that are much lower than the design capacities.
The EPA agrees that there are certain inherent throughput
limitations associated with the production of oil and natural
gas, primarily related to declining production rates. Therefore,
the final subpart HH specifies a method for calculating maximum
facility throughput to determine major source status and
applicability to subpart HH. This method is based on a
facility's past production rate and ability to document declining
annual operations. However, it is the responsibility of the
owner or operator to be aware of changes that could require a
facility to recalculate its PTE and to do so in a timely manner.
The owner or operator could be found in violation back until the
point in time at which an engineering judgement would have shown
that the facility was reasonably capable of emitting at major
source thresholds.
The EPA also received comments that the EPA should consider
the seasonal operation of natural gas storage facilities in
estimating potential emissions, and that the facility's PTE
cannot be based on withdrawal for the entire season at maximum
capacity. The commenters explained that natural gas storage
facilities must spend part of the year injecting gas, and that
withdrawal rates decrease as the storage field's pressure drops.
The EPA agrees that natural gas storage facilities have
inherent limitations due to the nature of their operations.
Therefore, the final rule (subpart HHH) contains a method for
calculating maximum facility throughput to determine major source
status and applicability of subpart HHH. The method is based on
the maximum withdrawal and injection rates and the working gas
capacity for a given storage field.
Several commenters recommended a simplified approach to
calculating PTE, such as screening equations similar to those
developed for other NESHAP, to take into account design and
operational limitations.
The EPA evaluated the use of an equation similar in
structure to the Gasoline Distribution NESHAP, 40 CFR part 63,
1-4
-------
subpart R. After extended effort, the EPA found that the number
of variables was too extensive to allow development of a
manageable equation.
Therefore, as an alternative, the EPA developed a simplified
major source determination (MSD) for HAP emission sources in the
oil and natural gas production and natural gas transmission and
storage source categories. The simplified MSD allows the owner
or operator of a facility to easily determine (1) if they are
major sources and whether NESHAP requirements apply to their
facility, and (2) if they are required to obtain a title V
operating permit.
The final subpart HH states that facilities, prior to the
point of custody transfer, that have a facility-wide actual annual
average natural gas throughput less than 18.4 thousand m3/day and
a facilitywide actual annual average hydrocarbon liquid
throughput less than 39,700 liter/day are exempt from subpart HH.
Owners and operators of production facilities, after the
point of custody transfer (including natural gas processing
plants), must aggregate emissions from all HAP emissions units at
the facility when determining whether or not the facility is a
major source. Production facilities, after the point of custody
transfer, are likely to have emission units in addition to glycol
dehydration units and storage vessels, such as amine treaters and
sulfur recovery units that are typically located at natural gas
processing plants. Since these emissions units must be included
in the total emissions for the facility, the EPA could not
develop a cutoff that would reasonably ensure that sources
operating below such a cutoff would not be major sources.
Therefore, production facilities located after the point of
custody transfer, including natural gas processing plants, do not
qualify for the simplified major source determination.
Using the same procedure, the EPA developed an MSD for
natural gas transmission and storage facilities where glycol
dehydration units are the only HAP emission points. The final
subpart HHH states that natural gas transmission and storage
facilities operating with an actual annual average natural gas
1-5
-------
throughput below 28.3 thousand m-Vday are exempt from
subpart HHH.
1.2.4 Averaging Periods
The proposed standards required a 95.0 percent control
efficiency for all control devices, but did not specify over
which averaging period the 95.0 percent should be determined. By
not specifying an averaging period, the proposed rule required
continuous compliance for all control devices. The EPA received
several comment letters requesting that the EPA specify an
averaging period. The commenters were particularly concerned
that condensers could not achieve a 95.0 percent control
efficiency on a continuous basis and that additional controls
would be required to ensure compliance with the 95.0 percent
requirement.
The commenters' primary point was that condensers are
significantly affected by changes in ambient temperature.
According to the commenters, when the ambient temperature is
high, the condensers are less efficient. The commenters were
concerned that during the warm summer months, condensers would
not meet the control requirements. Therefore, the commenters
specifically requested either a 30-day or a 12-month averaging
period for compliance with the control requirements to balance
changes in ambient temperature. In support of this request, the
commenters maintained that using a longer averaging period would
create no significant change in the emissions to the environment,
but would substantially decrease the number of technical
violations of the standard and reduce the administrative burden
for the industry and the EPA.
The EPA reviewed the control efficiency and averaging period
requirements in response to these comments. Based on the
Agency's review of the possible options, the final rules require
95.0 percent control as a daily average. As an alternative for
owners or operators that install condensers, the EPA has modified
subpart HH to allow 95.0 percent condenser control as a 365-day
rolling average, based on daily average condenser efficiency as a
function of condenser outlet temperature (i.e., at the end of
1-6
-------
each operating day, the owner or operator calculates the daily
average condenser outlet temperature, then calculates the 365-day
average control efficiency for the preceding 365 days, including
the current operating day).
Based on the information collected under the authority of
section 114 of the Act, the comments received during the public
comment period, and site visits, the EPA believes that an
averaging period shorter than 365 days is appropriate for the
natural gas transmission and storage source category. To the
Agency's knowledge, glycol dehydration units located at storage
facilities do not typically operate throughout the year.
Therefore, the EPA was concerned that it would take more than
I calendar year for a facility to obtain 365 days of data.
Additionally, glycol dehydration units located at these sources
do not typically operate during the warm summer months when
condenser efficiency is lower. Although transmission facilities
do operate for most of the year, the EPA believes that the HAP
emission units in operation at these facilities are primarily
compressors, and that most glycol dehydration units located at
these facilities are used for withdrawing natural gas from
storage (i.e., not likely to operate year-round). Therefore, for
condensers installed on glycol dehydration units subject to
control requirements under subpart HHH, the EPA has modified the
requirements to specify that owners or operators that install
condensers have the option of meeting a 95.0 percent control
efficiency as a 30-day rolling average.
1.2.5 process Modifications
Several commenters requested that the EPA allow for
combinations of controls and process modifications to achieve the
required control efficiency. The commenters provided several
suggestions for modifying the language in §63.765(c)(2) stating
that the owner or operator could reduce emissions from the glycol
dehydration unit by 95.0 percent through process modifications or
process modifications with controls.
The EPA agrees that owners or operators should be allowed to
achieve a 95.0 percent emission reduction using process
1-7
-------
modifications or combinations of process modifications and one or
more control devices. Therefore, the final rules contain
requirements for demonstrating compliance with a 95.0 percent
emission reduction using process modifications or a combination
of process modifications and one or more control devices. In
particular, the final rules require the owner or operator to
demonstrate how emissions have been reduced and to what level,
and that the facility continues to be operated such that the 95.0
percent emission reduction is maintained.
The final rules also require the owner or operator to
document facility operations and to provide this information in
the Periodic reports.
1.2.6 Standards for Natural Gas Transmission and Storage
The EPA received several comment letters expressing concern
for the EPA's proposed standard for the natural gas transmission
and storage source category. The commenters stated that the EPA
did not have sufficient data to develop standards for the natural
gas transmission and storage source category. The commenters
requested that the EPA delay the natural gas transmission and
storage portion of the proposed rulemaking to properly survey the
industry for more meaningful data and assess whether a standard
for the natural gas transmission and storage source category is
necessary or achievable.
Several commenters explained that a review of the background
information for proposed subpart HHH showed that the database
consisted of information on the methods used in natural gas
transmission from only two companies and no underground storage
facilities. The commenters noted that the companies surveyed
were predominately oil production facilities that handled gas as
a by-product of oil production and that have higher HAP emissions
because they handle more liquids with higher concentrations of
HAP.
In response to these comments, the EPA collected additional
data on glycol dehydration units in the natural gas transmission
and storage source category through site visits and requests for
information under the authority of section 114 of the Act.
1-8
-------
Through these site visits and survey questionnaires, the EPA
collected information from 83 facilities in the natural gas
transmission and storage source category. The EPA considered
this new information, along with the previously collected
information on the natural gas transmission and storage source
category, in developing a MACT floor for existing and new process
vents on glycol dehydration units located at facilities in this
source category. The EPA also used this information to better
characterize processes and operations at natural gas transmission
and storage facilities.
As stated in the January 15, 1999 supplemental notice (64 FR
2611), the additional data supported a MACT floor of 95.0 percent
for existing and new natural gas transmission and storage
facilities. In addition, the EPA announced that the Agency was
considering raising the proposed throughput cutoff of 85 thousand
m3/day to 283 thousand m3/day on an actual annual average basis.
Glycol dehydration units operating below this cutoff would not be
required to install controls under subpart HHH. The data did not
warrant a change in the benzene emission cutoff of 0.90 Mg/yr.
The public comment period closed on February 16, 1999. The
EPA received four comment letters in response to the EPA's
request for comments and supporting information on the
consideration of a 95.0 percent HAP emission reduction as the
floor level of control, on the 283 thousand m3/day natural gas
throughput cutoff and the 0.90-Mg/yr benzene emission cutoff.
The commenters agreed that exempting glycol dehydration units
with actual annual average natural gas throughputs less than
283 thousand 78 m3/day and with actual average benzene emissions
less than 0.90 Mg/yr from the control requirements under subpart
HHH was appropriate.
However, the commenters indicated that they did not agree
with a MACT floor of 95.0 percent for the transmission and
storage source category. The commenters requested that the final
rule should either exempt existing sources controlled by
condensers, or require that existing sources controlled with
condensers be controlled to a different level (i.e., 70 percent)
1-9
-------
than the combustion technology-based MACT floor. The commenters
stated that condensers could consistently achieve a 75 percent
emission reduction and that requiring an additional 20 percentage
points of emission reduction in HAP would be inconsistent with
the cost-to-benefit analysis in the February 6, 1998 proposal.
The EPA does not believe that it is necessary to provide
exemptions or alternative levels of control for existing glycol
dehydration units that are controlled by condensers. The EPA
believes that this would not be consistent with the Act, which
specifies in section 112(d)(3) that for a source category with 30
or more sources (such as the transmission and storage source
category), the MACT floor for existing sources shall not be less
stringent than "... the average limitation achieved by the best
performing 12 percent of the existing sources. . .." The data
collected by the EPA indicated that the average limitation
achieved by the top 12 percent of the existing glycol dehydration
units located at natural gas transmission and storage facilities
was 95.0 percent. Furthermore, the data indicated that the top
12 percent of the existing glycol dehydration units were
controlled using combustion or a combination of combustion and
condensation. Therefore, in accordance with the statute, the EPA
established the MACT floor to be 95.0 percent for glycol
dehydration units located at natural gas transmission and storage
facilities, which corresponds to combustion.
However, the EPA agrees that the supplemental notice did not
address the issue of averaging period for condensers in use at
transmission and storage facilities. As stated in this preamble,
the final rule allows an owner or operator that installs a
condenser for control of HAP from glycol dehydration unit process
vents to establish compliance with the 95.0 percent HAP emission
reduction on a 30-day rolling average. In addition, the final
rule allows the owner or operator to comply with one of the
following: (1) 95.0 percent HAP emission reduction, (2) 20 parts
per million by volume (ppmv) outlet HAP concentration for
combustion devices, or (3) outlet emissions of 0.90 Mg/yr of
benzene. The EPA believes that the 0.90 Mg/yr benzene emission
1-10
-------
limit and the 30-day averaging period for condensers provides
sufficient flexibility for owners and operators of existing
controlled glycol dehydration units.
1.2.7 Monitoring. Recordkeeping. and Reporting Requirements
The EPA received several comment letters claiming that the
recordkeeping and reporting requirements of the proposed rule
were extremely burdensome. The commenters requested that the EPA
reduce the monitoring, recordkeeping, and reporting burden
associated with the proposed rule. In particular, commenters
were concerned that remote and unmanned facilities would be
overburdened by the proposed monitoring, recordkeeping and
reporting requirements. Commenters also requested that
provisions be added to the rule to avoid duplicative reporting.
Other commenters requested that flexibility to allow alternative
monitoring, recordkeeping, and reporting be incorporated into the
final rule.
The EPA recognizes that unnecessary monitoring,
recordkeeping, and reporting requirements would burden both the
source and enforcement agencies. Prior to proposal, the EPA
attempted to reduce the amount of monitoring, recordkeeping, and
reporting to only that which is necessary to demonstrate
compliance.
Although the EPA has not removed the monitoring requirements
for unmanned or remote facilities, the EPA did evaluate the
possibility of reducing the requirements for unmanned facilities.
The EPA concluded, however, that the monitoring requirements are
the minimum necessary to ensure that control devices are
operating to ensure compliance.
The EPA reevaluated whether monitoring, recordkeeping, and
reporting requirements could be further reduced while maintaining
the enforceability of the rule. Therefore, the EPA has made the
following changes in the promulgated rule to further reduce the
monitoring, recordkeeping, and reporting burden.
(1) Almost all reports have been consolidated into the
Notification of Compliance Status report and the Periodic
reports.
1-11
-------
(2) If multiple tests are conducted for the same kind of
emission point, using the same test method, only one complete
test report is required to be submitted along with the summaries
of the results of other tests.
(3) Site-specific test plans describing quality assurance in
§63.7(c) of 40 CFR part 63, subpart A, are not specifically
required in the individual subparts because the test methods
cited in subparts HH and HHH already contain applicable quality
assurance protocols. It should be noted that the Administrator
would still have the authority to request a test plan.
(4) Periodic reports are required to be submitted
semiannually for all facilities (the proposal required quarterly
reports if monitored parameters were out of range more than a
specified percentage of time).
(5) A reduction in the record retention requirements for
monitored parameters. The proposal required values of monitored
parameters to be recorded every 15 minutes and all 15-minute
records had to be retained. The final rule requires monitored
parameters to be recorded every hour and all hourly records to be
retained.
1.3 SUMMARY OF IMPACTS OF PROMULGATED ACTION
The EPA estimated that the final oil and natural gas
production standards will reduce nationwide emissions of HAP by
approximately 30,000 megagrams per year (Mg/yr) from existing
sources, and 3,000 Mg/yr from new sources. The final natural gas
transmission and storage standards are estimated to reduce
nationwide HAP emissions by 390 Mg/yr from existing sources. No
new sources are anticipated for the natural gas transmission and
storage source category after the effective date for new sources
and in the first three years following promulgation of the
subpart HHH.
The nationwide annual costs (including capital recovery) of
the final rule are estimated to be approximately $4.0 million per
year for existing major sources in the oil and natural gas
production source category and $300,000 per year for existing
natural gas transmission and storage major sources. The total
1-12
-------
annual costs for new major oil and natural gas production sources
was estimated to be approximately $400,000 per year. The
economic analysis determined that the oil and natural gas
production regulation is anticipated to affect less than 5
percent of the total U.S. crude oil production, and thus, it is
unlikely to have any influence on the U.S. supply of crude oil or
world crude oil prices. In addition the imposition of regulatory
costs on the natural gas market has a negligible effect on
natural gas prices, output, employment, foreign trade, and
business profitability.
The secondary environmental impacts that occur as a result
of this rule are expected to be minimal in comparison to the
primary HAP reduction benefits from the implementation of the
control options. The rule encourages the use of emission
controls that recover hydrocarbon products (such as methane and
condensate) that can be used onsite for fuel or reprocessed for
sale.
The energy impacts associated with the operation of emission
control devices are not significant. The EPA estimated that the
annual energy requirements to be 38,000 kilowatt hours per year
and result from the operation of vapor collection and recovery
systems installed on storage vessels. The EPA estimated that
add-on control systems (e.g., condensers and flares) would not
require additional energy.
1-13
-------
2.0 SUMMARY OF PUBLIC COMMENTS
A total of 50 letters commenting on the proposed standard
and the background information document for the proposed standard
were received. A public hearing was not requested. A list of
commenters, their affiliations, and the EPA docket number
assigned to their correspondence is given in Table 2-1.
For the purpose of orderly presentation, the comments have
been categorized under the following topics:
1. APPLICABILITY
2. DEFINITIONS
3. ASSOCIATED EQUIPMENT
4. HAP EMISSION POINTS
5. IMPACTS
6. ECONOMIC ANALYSIS
7. LEGAL ISSUES [OTHER THAN ISSUES ASSOCIATED WITH THE EPA'S
INTERPRETATION OF SECTION 112(n)(4)(A) AND (B)]
8. PERMIT ISSUES
9. ENFORCEMENT ISSUES
10. CONTROLS
11. MONITORING, RECORDKEEPING, AND REPORTING
12. TEST METHODS
13. COMPLIANCE
14. WORDING OF REGULATIONS (OTHER THAN APPLICABILITY AND
DEFINITIONS)
15. GENERAL PROVISIONS
16. MISCELLANEOUS
17. GENERAL COMMENTS SPECIFIC TO SUBPART HHH (NOT OTHERWISE
ADDRESSED)
18. COMMENTS RECEIVED ON THE JANUARY 15, 1999 SUPPLEMENTAL
NOTICE (64 FR 2611)
2-1
-------
TABLE 2-1. LIST OF COMMENTERS ON PROPOSED STANDARDS. OF
PERFORMANCE FOR OIL AND NATURAL GAS PRODUCTION AND NATURAL GAS
TRANSMISSION AND STORAGE INDUSTRIES
Docket Item Number3 Commenter and Affiliation
IV-D-1 G. Von Bodungen
Louisiana Department of Environmental Quality
Office of Air Quality
P.O. Box 82135
Baton Rouge, Louisiana 70844
IV-D-2 G. Holliday
Holliday Environmental Services, Inc.
P.O. Box 2508
Bellaire, Texas 77402
IV-D-3 J. Henderson
TruTesT Analytical Consultants, Inc.
3500 North Causeway Boulevard, Suite 600
Metairie, Louisiana 70002
IV-D-4 R. Gow
Questar Corp.
P.O. Box 45433
Salt Lake City, Utah 84145
IV-D-5 T. LaSalle, HLP Engineering, Inc.
barryh@linknet.net (Via e-mail)
IV-D-6 S. Knis
The Dow Chemical Company
Midland, Michigan 48675
IV-D-7 V. Lajiness
The Coastal Corporation
500 Renaissance Center
Detroit, Michigan 48243
IV-D-8 W. Ebarb
Hi Trading and Transportation Group
IV-D-9 J. Matuszak
The Peoples Gas Light and Coke Company
130 East Randolph Drive
Chicago, Illinois 60601
IV-D-10 T. Hutchins
El Paso Energy Company
2-2
-------
TABLE 2-1 (continued)
Docket Item Number Commenter and Affiliation
IV-D-11 R- Metcalf
Louisiana Mid-Continent Oil and Gas
Association
801 North Boulevard, Suite 201
Baton Rouge, Louisiana 70802
IV-D-12 A. Evans
Consumers Energy Company
1945 West Parnall Road
Jackson, Mississippi 49201
IV-D-13 C. Reheis
Western States Petroleum Association
1115 llth Street, Suite 150
Sacramento, California 95814
IV-D-14 T. Horn
Harding Lawson Associates
202 Central SE, Suite 200
Albuqureque, New Mexico 87102
IV-D-15 J. Cantrell
Gas Processors Association
6526 East 60th Street
Tulsa, Oklahoma 74145
IV-D-16 B. Price
Phillips Petroleum Company
Bartelsville, Oklahoma 74004
IV-D-17 R. Taylor
True Oil Company
P.O. Drawer 2360
Casper, Wyoming 82602
IV-D-18 M. Wax
Institute of Clean Air Companies
1660 L Street NW, Suite 1100
Washington, District of Columbia 20036
IV-D-19 W. Airey
Vorys, Sater, Seymour, and Pease LLP
52 East Gay Street
P.O. Box 1008
Columbus, Ohio 43216
2-3
-------
TABLE 2-1 (continued)
Docket Item Number3 Commenter and Affiliation
IV-D-20 K. Beckett
Jackson & Kelly
1600 Laidley Tower
P.O. Box 553
Charleston, West Virginia 25322
IV-D-21 V. Ammirato
Columbia Gas Transmission
P.O. Box 1273
Charleston, West Virginia 25325
IV-D-22 R. Jones
American Petroleum Institute
1220 L Street, Northwest
Washington, District of Columbia 20005
IV-D-23 W. Flis
Exxon Company, U.S.A.
P.O. Box 2180
Houston, Texas 77252
IV-D-24 S. Waisley
U.S. Department of Energy
Washington, District of Columbia 20585
IV-D-25 D. McKinnon
Manufacturers of Emission Controls Association
1660 L Street, Northwest, Suite 1100
Washington, District of Columbia 20036
IV-D-26 W. Doyle
Marathon Oil Company
539 South Main Street
Findlay, Ohio 45840
IV-D-27 M. Atherton
Columbia Energy Group Service Corporation
12355 Sunrise Valley Drive, Suite 300
Reston, Virginia 20191
IV-D-28 C. Price
Chemical Manufacturers Association
1300 Wilson Boulevard
Arlington, Virginia 22209
IV-D-29 M. Chytilo
Environmental Defense Center
906 Garden Street
Santa Barbara, California 93101
2-4
-------
TABLE 2-1 (continued)
Docket Item Number3 Commenter and Affiliation
IV-D-30 A. Lee
Texaco, Inc.
P.O. Box 509
Beacon, New York 12508
IV-D-31 L. Beal
Interstate Natural Gas Association of America
L. Traweek, American Gas Association
(This comment letter contains a printing error
in the topical report, please see item IV-G-13
for the correction to this problem.)
IV-D-32 M. Lev-On
ARCO
444 S. Flower Street
Los Angeles, California 90071
IV-D-33 M. McThomas
Independent Oil and Gas Association of West
Virginia
P.O. Box 1791
Charleston, West Virginia 25326
IV-D-34 W. Sellars
Chevron U.S.A. Production Company
P.O. Box 1635
Houston, Texas 77251
IV-D-35 M. Blair
Colorado Department of Public Health and
Environment
4300 Cherry Creek Drive, South
Denver, Colorado 80246
IV-D-36 B. Mathur
Illinois Environmental Protection Agency
IV-D-37 B. Freeman
Shell E&P Technology Company
Bellaire Technology Center
P.O. Box 481
Houston, Texas 77001
IV-D-38 M. Fish
Enron Oil & Gas Company
P.O. Box 4362
Houston, Texas 77210
2-5
-------
TABLE 2-1 (continued)
Docket Item Number3 Commenter and Affiliation
IV-G-1 P. Cantle
Santa Barbara County (California) Air
Pollution Control District
26 Castilian Drive B-23
Goleta, California 93117
IV-G-2 J. Ives
Rocky Mountain Oil & Gas Association
1900 Grant Street, Suite 510
Denver, Colorado 80203
IV-G-3 C. Matthews
Interstate Oil and Gas Compact Commission
P.O. Box 53127
Oklahoma City, Oklahoma 73152
IV-G-5 R. White
TU Services, Inc.
1601 Bryan Street
Dallas, Texas 75201
IV-G-7 R. Jones
Dehy Condensers, Inc.
129 N. Glenwood Boulevard
Tyler, Texas 75702
IV-G-9 P. Bennett
KM Energy Inc.
One Allen Center
500 Dallas Street, Suite 500
Houston, Texas 77002
IV-G-11 B. Russell
Independent Petroleum Association of America
1101 Sixteenth Street, Northwest
Washington, District of Columbia 20036
IV-G-12 M. Fox
New Century Energies
P.O. Box 840
Denver, Colorado 80202
IV-G-13 L. Beal
Interstate Natural Gas Association of America
10 G Street, Northeast, Suite 700
Washington, District of Columbia 20002
(This document is a correction of the printing
error in item IV-D-31.)
IV-G-14 Unsigned/Concerned citizen
2-6
-------
TABLE 2-1 (continued)
Docket Item Number Commenter and Affiliation
IV-G-15 J. Courville
Louisiana Department of Environmental Quality
Air Quality Division
P.O. Box 82135
Baton Rouge, Louisiana 70884
IV-G-16 F. Dowling
Emission Testing Service, Inc.
P.O. Box 15075
Baton Rouge, Louisiana 70895
IV-G-17 J. Monfries
Metco Environmental
P.O. Box 598
Addison, Texas 75001
IV-G-36 Vincent D. Lajiness
Director, Environmental, Legislative, and
Regulatory Affairs
Coatal States Management
500 Renaissance Center
Detroit, MI 48243
IV-G-37 Mr. Philip Bennett
Manager, Government Affairs
KN Interstate Gas Transmission Co.
One Allen Center
500 Dallas Street, Suite 100
Houston, TX 77002
IV-G-38 Ms. Lisa Seal
Director, Environmental Affairs
Interstate Natural Gas Association of America
10 G Street, N.E. Suite 700
Washington, DC 20002
IV-G-39 Mr. Thomas D. Hutchins, P.E.
Director, Environmental, Health & Safety
El Paso Natural Gas Company
P.O. Box 1492
El Paso, TX 79978-1492
a The docket number for this project is A-94-04.Dockets are on
file at the EPA Headquarters in Washington, D.C.
2-7
-------
2.1 APPLICABILITY
2.1.1 Determination of Manor Source Status
Comment: Commenters IV-D-11, IV-D-13, IV-D-14, IV-D-15,
IV-D-19, IV-D-22, IV-D-34, IV-D-37, IV-G-03, and IV-G-05
suggested that any source currently covered by a Federally,
State, or otherwise enforceable limit (e.g., title V permit)
should be able to include the control efficiency of the control
device when calculating applicability to subparts HH and HHH.
Commenter IV-D-11 recommended that the EPA exempt facilities
that were determined to be minor sources under part 70 from the
major source definition. The commenter stated that not excluding
"controlled" sources from the major source definition is
inconsistent with the intent of section 112(a)(1) of the CAA.
The commenter explained that several sources in Louisiana have
applied federally enforceable controls well before the date of
the proposal and that being considered "minor sources" under part
70, but "major sources" under this proposal is inconsistent for
these sources. The commenter stated that the two programs must
have identical interpretations of the term. According to the
commenter, Louisiana sources would be penalized for reducing
emissions several years before the proposal of the rule, which
sends the wrong signal to the regulated community.
Commenter IV-D-34 requested that the EPA specifically
require in §§63.760 and 63.771(d) that the potential to emit
(PTE) for an affected source be determined "considering all
controls and limitations at the source." Commenters IV-D-13 and
IV-D-37 stated that the following must be assumed:
• the control device was installed before the enforcement
date of the final national emission standards for
hazardous air pollutants (NESHAP);
• the control device was installed pursuant to a State or
local air quality law, ordinance, rule, requirement or
company business practice that was in place before the
enforcement date of the final NESHAP; and
• the operation and emission reductions achieved by the
control device are federally enforceable through a
facility's title V permit or through another means that
would ensure federal enforceability.
2-8
-------
According to the comraenters, their proposed criteria provide a
common sense method to calculate PTE for existing facilities that
have existing control devices and that have achieved early
emission reductions before the enforcement date of the final
NESHAP. The commenters further stated that their proposed
criteria also prevent a facility operator from avoiding the
intent of title III of the Clean Air Act (CAA).
Response; Facilities with HAP emissions equal to or greater
than the major source levels as established in the CAAA of 1990
are subject to the major source provisions of subpart HH. As
defined in §63.2 (subpart A), PTE estimates take into account
those controls installed due to regulatory requirements of
Federally-enforceable programs, which are defined in §63.2 and
the part 70 permit programs. Therefore, facilities with
controlled HAP emissions less than the major source thresholds
would be considered area sources. Therefore, by referring to
§63.2 of subpart A (see Table 2), subparts HH and HHH already
contain the provisions requested by the commenters.
For additional information on limiting PTE for section 112
purposes and for other reasons, please refer to the following
memoranda: (1) January 25, 1995 Memorandum from John Seitz,
Director, OAQPS, entitled "Options for Limiting the Potential to
Emit (PTE) of a Stationary Source Under Section 112 and Title V
of the Clean Air Act;" (2) August 27, 1996 Memorandum from John
Seitz, Director, OAQPS, entitled "Extension of January 25, 1995
Potential to Emit Transition Policy;" and (3) July 10, 1998
Memorandum from John Seitz, Director, OAQPS, entitled "Second
Extension of January 25, 1995 Potential to Emit Transition Policy
and Clarification of Interim Policy."
Comment: Commenter IV-D-19 pointed to a court case
(National Mining Congress v. EPA. a59 F.3d.1351, D.C. Cir.1995)
2-9
-------
where the District of Columbia (DC) Circuit Court of Appeals
ruled that the EPA had not adequately justified the requirement
in section 112 of the CAA that standards that place limits on PTE
must be "federally enforceable." The cotnmenter suggested that
within this rulemaking, limits on PTE are not limited to those
that are "federally enforceable" as stated in the General
Provisions (40 CFR 63.2). According to the cotnmenter, any
physical or operational limitation on the capacity of a source to
emit a pollutant is appropriate. The commenter also suggested
that the effect a limitation would have on emissions should be
either federally enforceable or legally and practically
enforceable by the State.
Response; In the National Mining court case, the court
required the EPA to reconsider the Federal enforceability
requirement, but did not vacate the requirement. As a result,
the requirement for federal enforceability is still in effect.
The definition of PTE for the MACT program (40 CFR 63.2) is
currently under review and the EPA is engaged in a rulemaking
process to amend the requirements in the General Provisions.
Therefore, the EPA has not modified subparts HH and HHH in
response to this comment.
Comment: Commenters IV-D-08, IV-D-15, IV-D-17, IV-D-20,
IV-D-22, IV-D-23, IV-D-34, IV-G-03, and IV-G-05 were concerned
that PTE estimates, as defined in the General Provisions, would
be unrealistically high and would subject many small
insignificant sources to the maximum achievable control
technology (MACT) requirements. Commenter IV-G-03 was concerned
that the PTE calculations this would result in high costs for
controlling low emission sources, and may force marginally
economic wells into premature abandonment. These commenters,
along with commenters IV-D-04, IV-D-13, IV-D-19, IV-D-26,
IV-D-30, IV-D-31, and IV-G-11 requested that PTE be based on the
inherent design and operational limitations of production and
transmission and storage facilities, such as throughput rates.
2-10
-------
Commenter IV-D-04 suggested that Gas Research Institute
(GRI)--GLYCalc™, Version 3.0 or higher (GLYCalc) would allow for
the inclusion of these operating conditions in determining
applicability for affected sources.
According to commenters IV-D-15, IV-D-22, IV-D-31, IV-D-34,
IV-G-02, IV-G-03, and IV-G-05, the throughput of oil and gas
production operations decline over time, and existing equipment
is often designed, constructed and operated based on high initial
production rates. Therefore, the commenters suggested that the
facilities are usually operated at actual throughput rates that
are much lower than the design capacities. Commenter IV-D-15
remarked that the throughput or process rate of a unit is limited
by the oil or gas available in the geographic area where it is
located. The commenter explained that the product being handled
has unique chemical characteristics such as American Petroleum
Institute (API) gravity, gas-to-oil ratio (GOR), etc., which also
affect the emissions from a unit. The commenter further
explained that the other equipment at the site will also affect
the potential emissions of a unit (i.e., capping the potential
emissions).
According to commenters IV-D-05, IV-D-15, IV-D-22, IV-D-34,
IV-G-03, and IV-G-05, several States have established, through
their own permit programs, mechanisms to limit PTE. The
commenters requested that this MACT defer to State programs. The
commenters suggested that the methodology in the Texas Natural
Resource Conservation Commission's (TNRCC) Oil and Gas
Supplemental Guidance Memorandum be used to define the PTE at
inherently limited sources. According to commenter IV-D-15, to
calculate PTE using the TNRCC approach, the operator (1) averages
the highest site product throughput over the past five years, (2)
multiplies that average by 1.2 (raising the throughput 20 percent
covers the possibility if minor fluctuations or changes), and (3)
uses the highest impact chemical composition from the past five
years. Texas also defines calculation methods for inherently
limited emission units and documentation, monitoring and
recordkeeping requirements. Commenter IV-D-34 also noted that
2-11
-------
Wyoming allows exploration and production facilities to adopt
design or other limitations under State regulations.
Commenter IV-D-22 also provided supplemental comments
(IV-G-23) and recommended that the EPA either (1) specify methods
for the oil and natural gas production source category to use in
calculating PTE or (2) provide a simple federal synthetic minor
source mechanism.
Response: The EPA agrees that there are certain inherent
throughput limitations associated with the production of oil and
natural gas, primarily related to declining production rates.
Therefore, the EPA has developed an approach for determining
whether or not a facility is a major source subject to
subpart HH. The final rule allows an owner or operator to
calculate potential emissions using a maximum annual facility
throughput that is calculated as follows:
1. If the owner or operator of a production facility documents,
to the Administrator's satisfaction, a decline in annual
natural gas or hydrocarbon liquid throughput, each year, for
the five years prior to the effective date of subpart HH,
the owner or operator must determine the maximum natural gas
or hydrocarbon liquid throughput as the average of the
annual natural gas or hydrocarbon liquid throughput for the
three years prior to the effective date of subpart HH,
multiplied by 1.2. This maximum throughput must be used to
determine a facility's PTE.
2. If the owner or operator cannot document a decline in annual
throughput each year for the five years prior to the
effective date of subpart HH, the maximum throughput used to
calculate PTE must be calculated as the highest annual
natural gas or hydrocarbon liquid throughput over the five
years prior to the effective date of subpart HH, multiplied
by a factor of 1.2.
3. The owner or operator is required to document annual
facility natural gas or hydrocarbon liquid throughput each
year and if the facility's natural gas or hydrocarbon liquid
throughput increases above the maximum throughput calculated
2-12
-------
in steps (1) or (2), the maximum throughput must be
recalculated using the new, higher, throughput multiplied by
the factor of 1.2.
4. The owner or operator is also required to determine the
maximum values for other parameters used to calculate PTE as
the maximum for the period over which the maximum natural
gas or hydrocarbon liquid throughput is determined in steps
(1) or (2).
Comment: Commenters IV-D-07 and IV-D-31 requested that the
EPA consider the seasonal operation of natural gas storage
facilities in estimating potential emissions and that the
facility's PTE cannot be based on withdrawal for the entire
season at maximum capacity. The commenters explained that
natural gas storage facilities must spend part of the year
injecting gas and that withdrawal rates decrease as the storage
field's pressure drops.
Response; The EPA agrees that natural gas storage
facilities have inherent limitations due to the nature of their
operations. Information collected during site visits indicated
that glycol dehydration units located at storage facilities
normally operate in the winter when gas is being withdrawn from
storage fields (Air Docket A-94-04 numbers IV-B-01 through
IV-B-05). Therefore, the EPA believes that it is not appropriate
for such facilities to estimate potential emissions based on
year-round operation (i.e., 8,760 hr/yr). Therefore the EPA has
developed the following procedure to determine major source
status and applicability to subpart HHH for facilities that store
natural gas or facilities that transport and store natural gas:
1. The owner or operator calculates the number of hours it
takes to complete a storage cycle for the facility. The
storage cycle is the number of hours for the injection
cycle, calculated using Equation 1, plus the number of hours
for the withdrawal cycle, calculated using Equation 2.
2-13
-------
1C =
WGC
(1)
where:
1C
WGC
Facility injection cycle in hr/cycle.
Working gas capacity in m3. The working
gas capacity is defined as the maximum
storage capacity minus the FERC
cushion.1
IRmax
Maximum facility injection rate in
m3/hr.
WC =
WGC
(2;
where:
WC
WGC
Facility withdrawal cycle in hr/cycle.
Working gas capacity in m3 (same value
used in equation 1) .
Maximum facility withdrawal rate in
m3/hr.
2. The owner or operator calculates the number of storage
cycles per year using Equation 3.
_ 7 8760 hr/yr
Cycle - —
1C + WC
(3)
where:
Cycle
Number of storage cycles for the
facility per year (cycle/facility/yr)
1The FERC cushion is the minimum gas capacity allowed for a
storage field, as regulated by the Federal Energy Regulatory
Commission.
2-14
-------
1C = Number of hours for a facility injection
cycle, calculated using Equation 1
(hr/cycle).
WC = Number of hours for a facility
withdrawal cycle, calculated using
Equation 2 (hr/cycle).
3. The owner or operator calculates the facilitywide maximum
annual glycol dehydration unit hours of operation calculated
using Equation 4.
Operation = Cycles * WC (4)
where:
Operation = Facilitywide maximum annual glycol
dehydration unit hours of operation
(hr/yr).
Cycles = Number of storage cycles for the
facility per year, calculated in
Equation 3 (cycle/facility/yr).
WC = Number of hours for a facility
withdrawal cycle (hr/cycle) as
calculated in Equation 2.
4. The owner or operator calculates the maximum facilitywide
natural gas throughput using Equation 5.
Throughput - Operation x (V^max (5)
where:
Throughput = Maximum facilitywide natural gas
throughput in m^/yr.
Operation = Maximum facilitywide annual glycol
dehydration unit hours of operation
in hr/yr, as calculated in Equation
4.
= Maximum facility withdrawal rate in
m3/hr.
2-15
-------
Since transmission facilities do not spend part of the year
injecting gas into storage, the EPA believes that the approach
for storage facilities is not appropriate. Therefore, the EPA
has included different requirements in subpart HHH for these
facilities to account for year-round operation. For facilities
that only transport natural gas, the final subpart HHH requires
owners or operators to calculate the maximum facility natural gas
throughput as the highest annual natural gas throughput over the
five years prior to the effective date of the rule, multiplied by
1.2.
The final subpart HHH also contains requirements for
determining maximum values for other parameters used to
calculated potential emissions and for documenting annual
facility natural gas throughput. These requirements are the same
as those specified for production.
Comment: Commenters IV-D-22, IV-D-26, IV-D-30, IV-D-34,
IV-G-02, and IV-G-11 recommended a simplified approach to
calculating PTE, such as screening equations similar to those
developed for other NESHAP, to take into account design and
operational limitations (e.g., Gasoline Distribution, 40 CFR
part 63, subpart R).
Commenter IV-D-26 mentioned the possibility of a source
category-specific definition for PTE. Commenter IV-G-02 stated
that a simplified PTE analysis should be available for
determining applicability to subparts HH and HHH for the
following reasons:
• oil and gas equipment may be oversized compared with
its available throughput (due to field depletion or
future field development plans),
• operators must make decisions on several sources, and
• the EPA's definition of what must be aggregated in
subparts HH and HHH for a major source determination
can be different from the basis for major source
determination for title V (63 FR 6300 - 6303 cited).
2-16
-------
Commenter IV-D-22 recommended that §63.760 (c) be amended to
include the following text:
(1) The owner or operator of an affected source may demonstrate
that the source is an area source for all purposes under
this subpart by documenting and recording as required that
either:
(i) [Reserved for screening equations, e.g., the result "x"
of the following equation is "y"]; or
(ii) Specific operational or physical limitations adopted
for the source result in an area source classification.
Such limitations may include (1) parameters of the
hydrocarbon fluid, (2) operating/production parameters
of the facility, (3) parameters of the hydrocarbon
reservoir, and (4) any other reasonable and enforceable
parameter.
(2) An area source classification established pursuant to these
criteria shall be treated as part of the design of the
source if it is federally, state, or otherwise practically
enforceable.
The commenter recommended that the EPA create a process within
subpart HH that streamlines the specification of enforceable
applicability criteria as referenced in modified §63.760(c)(ii),
above. The commenter stated that they will submit supplemental
comments providing appropriate criteria and outlining an
appropriate methodology for establishing this process. In their
supplemental comments (Air Docket A-94-04, number IV-G-23), the
commenter recommended a screening process for determining major
source status. This process included steps to (1) evaluated the
source status of glycol dehydration units, (2) evaluate the
source status for storage vessels, and (3) evaluate the source
status of collocated equipment. The commenter also made
recommendations monitoring, recordkeeping and reporting
requirements.
Response; The EPA evaluated the use of an equation similar
in structure to the Gasoline Distribution NESHAP, 40 CFR part 63,
subpart R. After extended effort, the EPA found that the number
of variables was too extensive to allow development of a
manageable equation.
2-17
-------
Therefore, as an alternative, the EPA has developed a
simplified major source determination (MSD) for HAP emission
sources in the oil and natural gas production and natural gas
transmission and storage source categories, in addition to the
PTE approach outlined in a previous comment. The simplified MSD
allows the owner or operator of a facility to easily determine
(1) if they are major sources and whether MACT requirements apply
to their facility, and (2) if they are required to obtain a title
V operating permit.
The objective of the simplified MSD is to set applicability
thresholds that would reasonably ensure that no facilities
operating below such a threshold would have HAP emissions greater
than the major source thresholds of 10 tpy for individual HAP and
25 tpy for any combination of HAP as defined in the CAA. A
detailed description of the development of this MSD is presented
in the docket (Air Docket A-94-04 number IV-A-12).
To develop this MSD, the EPA reviewed "reasonable worst
case" scenarios for use in development of the simplified MSD
applicability levels. These "reasonable worst case" scenarios
take into account such variables as throughput, HAP
concentrations, and standard operating procedures.
Based on these scenarios, the EPA determined that oil and
natural gas production facilities prior to the point of custody
transfer, with a facilitywide actual annual average natural gas
throughput less than 650 thousand standard cubic feet per day
(scf/d) can be reasonable expected not to exceed the major source
thresholds. Likewise, the EPA determined that oil and natural
gas production facilities prior to the point of custody transfer
with a facilitywide hydrocarbon liquid throughput less than 250
2-18
-------
bpd can reasonably be expected not to exceed the major source
thresholds.
Storage vessels make up a small percentage of emissions from
a production facility and have different emission profiles as
compared to glycol dehydration units. Therefore, the EPA
determined that a production facility consisting of glycol
dehydration units and storage vessels that meet the 650
thousand-scf/d natural gas throughput and the 250-bpd hydrocarbon
liquid throughput thresholds can be expected not to exceed the
major source thresholds.
Section 63.760 of final subpart HH contains an exemption
that states that production facilities prior to the point of
custody transfer, with a facilitywide natural gas throughput less
than 650 thousand scf/d and a facilitywide hydrocarbon liquid
throughput less than 250 bpd are exempt from subpart HH.
Owners and operators of production facilities after the
point of custody transfer (including natural gas processing
plants) are required to aggregate emissions from all HAP
emissions units at the facility when determining whether or not a
facility is a major source. Furthermore, production facilities
after the point of custody transfer are likely to have other HAP
emission units in addition to glycol dehydration units and
storage vessels, such as amine treaters and sulfur recovery units
which are typically located at natural gas processing plants.
Since emissions from these emission points must be aggregated in
determining the major source status of the facility, the EPA
determined that it would be unreasonable to develop a throughput
cutoff that would reasonably ensure that facilities operating
below such cutoff would not be a major source. Therefore,
production facilities located after the point of custody
2-19
-------
transfer, including natural gas processing plants, do not qualify
for the simplified MSB.
Using this same approach, the EPA determined that natural
gas transmission and storage facilities with a facilitywide
actual annual average natural gas throughput less than 1 million
standard cubic feet per day (MMscf/d), can reasonably be expected
not to exceed the major source thresholds (Air Docket A-94-04
number IV-A-15). Section 63.1270 of final subpart HHH states
that facilities operating with an actual average annual natural
gas throughput less than 1 MMscf/d are exempt from subpart HHH.
However, since owners or operators of facilities in the natural
gas transmission and storage source category must aggregate
emissions from all HAP emissions units to determine major source
status, this exemption only applies to facilities where glycol
dehydration units are the only HAP emissions unit.
Comment: Commenter IV-D-26 recommended using the logic in
the EPA's 1995 Potential to Emit Transition Policy. Under this
policy, sources with low emissions (e.g., less than 50 percent of
major source thresholds) may be deemed nonmajor if records of
actual emissions are kept. Commenters IV-D-08, IV-D-20, and
IV-D-22 suggested the use of written documentation of physical
and operational limitations that would be federally, State, or
otherwise practically enforceable. The commenters recommended
that the EPA provide operators the ability to select maximum
annual levels for product throughput, and continuous maximums for
physical parameters of the product received and operating
parameters under which the unit will be operated. The operator
would then calculate PTE based on these maximums using accepted
calculation procedures (e.g., Vasquez-Beggs, or GLYCalc) and MACT
would apply if the aggregate PTE calculated based on maximums
exceeds the major source thresholds.
2-20
-------
Response; In the January 25, 1995 policy memorandum
entitled "Options for Limiting the Potential to Emit (PTE) of a
Stationary Source Under Section 112 and Title V of the Clean Air
Act (Act)," the EPA issued a transition policy for section 112
and title V (this memorandum is available on the EPA's web site
at Internet address http://www.epa.gov/ttn/oarpg/t5pgm.html) .
This transition policy addressed concerns that some sources may
face gaps in the ability to acquire federally enforceable PTE
limits because of delays in State adoption of EPA approval of
programs or in their implementation. In order to ensure that
such gaps would not create adverse consequences for States or for
sources, the EPA provided that during a 2-year period extending
from January 1995 through January 1997, for sources lacking
federally enforceable limitations, State and local air regulators
had the option of treating the following types of sources as
non-major under section 112 and in their title V programs:
1. sources that maintain adequate records to demonstrate
that their actual emissions are less than 50 percent of
the applicable major source threshold, and have
continued to operate at less than 50 percent of the
threshold since January 1994, and
2. sources with actual emissions between 50 and 100
percent of the major source threshold, but which hold
State-enforceable limits that are enforceable as a
practical matter.
On August 27, 1996, this transition policy was extended until
July 31, 1998 (Internet site
http://www.epa.gov/ttn/oarpg/t5pgm.html). On July 10, 1998, in a
memorandum entitled "Second Extension of January 25, 1995
Potential to Emit Transition Policy and Clarification of Interim
Policy" (Internet site http://www.epa.gov/ttn/oarpg/t5pgm.html),
the EPA announced a second extension of the transition policy.
2-21
-------
These extensions were provided because the EPA is engaged in a
rulemaking process to consider amendments to the current PTE
requirements. Currently, the PTE rulemaking, which will address
the PTE requirements in the General Provisions (40 CFR part 63,
subpart A) and the title V operating permits program, has not
been completed. These rule amendments will affect federal
enforceability requirements for PTE limits under these programs.
Thus, there will continue to be uncertainty with respect to
federally enforceable limits. Therefore, in the July 10, 1998,
the EPA extended the transition policy until December 31, 1999,
or until the effective date of the final rule in the PTE
rulemaking, whichever is sooner.
The EPA expects that the rulemaking will be completed before
December 31, 1999, and owners or operators will have the option
of complying with the PTE rulemaking as well as the procedures
specified in subparts HH and HHH.
2.1.2 Exemptions
Black Oil
Comment: Commenters IV-D-17 and IV-D-24 were concerned
about the exemption criteria for facilities that process, store,
or transfer black oil. Commenter IV-D-24 supported the use of a
black oil exemption in the proposed standards. Commenter IV-D-17
suggested that pipelines that transmit "black oil" should not be
further considered a potential HAP source.
. Response; As stated in the preamble to the proposed rule,
pipelines that handle hydrocarbon liquids after the point of
custody transfer are not within the scope of the oil and natural
gas production source category (63 FR 6291). The EPA plans to
define the organic liquids distribution (non-gasoline) source
category to include those facilities that distribute hydrocarbon
liquids after the point of custody transfer. Since black oil is
2-22
-------
defined as a hydrocarbon liquid, facilities that transmit black
oil after the point of custody transfer will be covered under the
organic liquids distribution NESHAP. The EPA does not believe
that addressing this issue within subpart HH is necessary.
Comment: Commenters IV-D-20 and IV-D-33 questioned the
EPA's basis for the definition of black oil in subpart HH.
Commenter IV-D-20 stated that it was unclear whether this
definition was based upon an assessment of HAP emissions or upon
the determination that black oil that meets this definition in
certain quantities and stored in a specific manner would result
in adverse impact upon human health and/or the environment.
Commenters IV-D-12, IV-D-33, and IV-D-38 requested changes
to the GOR and API gravity cutoffs proposed in the definition of
black oil in subpart HH. To be consistent with industry
practice, commenter IV-D-12 requested that the definition of
black oil be revised to a GOR of less than 5,000 standard cubic
feet per barrel (scf/bbl) and commenters IV-D-12 and IV-D-38
requested an API gravity less than 50°. Commenter IV-D-33
requested that the threshold be changed from an API gravity of
40° to 45°, which would provide additional regulatory relief to
producers already hindered by marginal production in the
Appalachian region. According to the commenter, Appalachian
Basin crude oil runs between 40 and 45°. [Note: The commenter
had 45° instead of 40° as the proposed specific gravity
threshold. A typographical error is likely.]
Response; During the development of proposed subpart HH,
industry representatives stressed that their industry was
composed of large numbers of facilities that handle black oil and
that black oil was not a significant contributor to overall
source category HAP emissions. The EPA evaluated the available
information and agreed that facilities that process black oil
were not significant sources of overall HAP emissions from the
2-23
-------
source category. Therefore, the EPA developed an exemption for
facilities that exclusively process, handle and store black oil.
Furthermore, the EPA did not identify control technologies
designed to reduce HAP in use at existing facilities that
exclusively process, handle, or store black oil. Therefore, the
EPA determined that the MACT floor was no control. This
determination was not made based on health risks associated with
black oil.
The EPA developed the definition for black oil (Air Docket
A-94-04 number IV-A-05) based on a series of articles by William
D. McCain (primary author).2'3 According to the information in
these articles, five types of reservoir fluids exist: black oil,
volatile oil, retrograde gas-condensate, wet gas, and dry gas.
Of these, black oils and volatile oils exist as liquids in the
reservoir. Black oil, which is a mixture of chemical species
ranging from methane to large, heavy, nonvolatile molecules, is
in solution with dry gas, which is primarily methane.4 Volatile
oils, which contain fewer heavy molecules, are in solution with
retrograde gas, which has fewer of the heavy organic molecules.
Reservoir fluid types are indicated by rules of thumb based
on initial producing GOR, stock-tank liquid gravity, and
stock-tank liquid color. Fluid type is usually determined by
initial producing GOR and can be confirmed using stock-tank
2McCain, William D. "Heavy Components Control Reservoir
Fluid Behavior." Journal of Petroleum Technology. September
1994. pages 746-750.
3McCain, William D. "Black Oils and Volatile Oils - What's
the Difference." Petroleum Engineer International. November
1993. pages 24-27.
4 Reference 2.
2-24
-------
gravity and color. [Note: The distinction between initial
producing GOR and producing GOR is important. As reservoir
pressure decreases over time, the producing GOR for black oil
increases. Therefore, if any other GOR is used, the facility may
not appear to qualify for the exemption.] The rule-of-thumb for
volatile oils is an initial producing GOR of at least
1,750 scf/bbl. Volatile oil is also suspected with a gravity of
40° or more and a color that is brown, reddish, orange, or green.
The rule-of-thumb for black oil is an initial producing GOR less
than 1,750 scf/bbl and an API gravity less than 45° and a color
that is dark, usually black (sometimes with a greenish cast) or
brown.
The EPA used the descriptions of black oil from these
articles to develop the proposed definition of black oil. Since
color determination is subjective, the EPA selected initial
producing GOR and API gravity as quantifiable criteria for
defining black oil. In addition, since the API gravity criteria
overlap for black oil and volatile oil, the EPA chose the lower,
more conservative value of 40° for the black oil definition. The
EPA believes that using a higher API gravity to define black oil,
such as 45 or 50° as recommended by the commenters, would
increase the possibility that the liquid is a volatile oil, thus
exempting sources that are likely to have higher HAP emissions.
The criteria for defining black oil, which were obtained directly
from widely recognized definitions of black oil and volatile oil
that are used in the oil and natural gas industry, are
technically sound for identifying which sources are included as
black oil facilities. Therefore, the EPA has not made any
changes to the definition of black oil in response to these
comments.
2-25
-------
Comment: Commenters IV-D-01 and IV-D-29 were concerned that
the exemption criteria would exempt facilities with significant
emissions. Commenter IV-D-01 requested that the EPA delete the
provision exempting black oil facilities from the requirements of
the subpart [§63.760 (e)]. According to the commenter, most oil
and gas production facilities in Louisiana would probably be
exempt from the subpart. Furthermore, the commenter stated that
oil with an API gravity of 40 degrees is light crude and is
almost condensate. The commenter also stated that oil with a GOR
of 1,750 scf/bbl would be expected to result in high gas
production.
Commenter IV-D-29 supported lowering the black oil
applicability thresholds from a gas-to-oil ratio (GOR) less than
1,750 scf/bbl and an API gravity less than 40° to a GOR of less
than 1,250 scf/bbl and an API gravity less than 27 or 28°. The
commenter was concerned that the proposed applicability
thresholds for black oil would exempt nearly all tank batteries
in Santa Barbara County, California (diesel fuel has an API
gravity of 38°).
Response; Based on an evaluation of the available
information, the EPA determined that there is a low potential for
HAP emissions from black oil in the oil and natural gas
production source category. The top 12 percent of facilities in
this subcategory were not controlled, and due to the low
emissions potential, it was determined to be not cost effective
to go beyond the MACT floor. Therefore, the EPA established an
exemption from regulatory requirements in subpart HH for those
facilities that exclusively handle black oil.
Furthermore, based on the EPA's understanding of the
characteristics of black oil, there may be significant gas
production from facilities that exclusively handle black oil.
However, this gas would primarily have a low moisture content,
and generally have a low potential for HAP emissions. Therefore,
2-26
-------
the EPA believes that facilities that process, store or handle
black oil are not significant sources of HAP emissions and has
not made any changes to the black oil definition in response to
this comment. The EPA believes that the proposed applicability
cutoffs are appropriate.
Comment: Commenter IV-D-02 recommended that the EPA
eliminate the definition for black oil in subpart HH, and "define
oil for purposes of part 63 as liquid hydrocarbons as Mineral
Management Service (MMS) does (30 CFR §206.51)." The commenter
agreed with the proposal to exempt black oil facilities provided
the definition of black oil was correct. According to the
commenter, defining black oil, which is dependant on many
variables, makes subpart HH too complex, and makes enforcement
impossible. The commenter stated that the EPA's definition of
"black oil" is arbitrary and capricious, and "totally neglects
long established and technically supportable definitions of
condensate and oil." The commenter noted that the EPA did not
include a discussion on reservoir condition of the hydrocarbon.
The commenter stated that although the EPA was correct in
dividing hydrocarbons into two categories (black oil and
condensate) , industry and MMS divide hydrocarbons differently
into "oil" and "condensate." The commenter suggested that
imposing two conflicting definitions of condensate and oil will
result in unwarranted confusion within industry and the agency.
The commenter felt that the EPA's approach to define "black oil"
as something different from "oil" is not technically correct and
is confusing.
Response; As stated in a previous response, the definition
of black oil was developed using industry-defined terms. The EPA
believes that the gas that evolves from black oil does not
contain significant amounts of HAP. Therefore, the distinction
between black oil and volatile oil is important. The commenter's
proposed oil definition does not distinguish between volatile oil
2-27
-------
and black oil. Therefore the EPA believes that a black oil
definition based on the MMS definition of oil would exempt
sources with significant HAP emissions.
Comment: Comtnenter IV-D-16 recommended that §63.760 (e) be
amended to allow for the production of 10 thousand cubic feet per
day (MCF/D) of casing head gas for facilities that are otherwise
subject to the black oil exemption. The commenter explained that
most oil production facilities that process "black oil" produce a
small amount of "casing-head gas." The commenter defined
"casing-head gas" as a gas dissolved in the oil that separates
from the oil as production occurs. According to the commenter,
the casing-head gas produced by a black oil facility is not
economically significant, but is a by-product of the oil
production process.
Response; Instead of specifying casing head gas as being
allowed, the EPA believes that any gas brought on site for fuel
or gas generated from black oil should be allowed at a black oil
facility. Therefore, §63.760(e)(1) states that for subpart HH,
"...a black oil facility that uses natural gas for fuel, or
generates gas from black oil..." is still exempt.
Glycol Dehydration Units
Comment: Several commenters referred to the flowrate and
benzene emission rate exemptions for glycol dehydration units.
Commenter IV-D-07 requested guidance on determining the annual
average for dehydrator de minimis and recommended that the
guidance be provided in §63.772(b). Commenters IV-D-24 and
IV-D-35 supported the use of flowrate and benzene emission rate
exemptions as it focuses on higher emissions, and according to
commenter IV-D-35, triethylene glycol (TEG) units are usually
located only at area -sources. Commenter IV-D-12 requested that
the EPA clarify the methods proposed for determining dehydrator
HAP emission-based applicability and that the EPA provide
examples to show how these methods should be applied.
2-28
-------
Response; In response to several comments the EPA has made
changes to final subparts HH and HHH to clarify the compliance
demonstration requirements (see section 2.14 for further
discussion). In addition, §§63.772(b) of final subpart HH and
63.1282(a) of final subpart HHH specify how the average natural
gas flowrate is to be calculated. The final rules specify that
emissions must be determined based on representative operations
and the EPA believes that the owner or operator should have
records for the representative operation of each glycol
dehydration unit. The EPA will be publishing implementation
guidance following promulgation of subparts HH and HHH.
Comment: Commenter IV-D-29 stated that they support the
following:
(I) lowering the natural gas applicability threshold for glycol
dehydration units from 85 thousand m3- to 42 thousand m3 .
The commenter stated that the EPA offered no real
justification for the selected applicability threshold,
(2) lowering the benzene emission applicability threshold for
glycol dehydration units from 0.9 ton per year to 0.5 ton
per year. The commenter stated that the potential health
effects of benzene exposure and the significance of total
HAP emissions from the source category justify this change,
(3) replacing "or" with "and" when discussing glycol dehydration
unit applicability thresholds. Thus, only those units that
meet both the natural gas throughput and benzene emission
rate would be exempted from the 95-percent control level,
and
(4) establishing control measures for those glycol dehydration
units that do meet all the applicability thresholds.
Response; The EPA evaluated several options in attempting
to establish applicability criteria for glycol dehydration units.
These options included a series of throughput, benzene emission
rates, and the use of the term "or" or "and" within the
applicability criteria. Based on its evaluation and to exempt
those emission points with low HAP emissions, the EPA does not
2-29
-------
believe that changing its applicability criteria for glycol
dehydration units is necessary. Furthermore, there was no
evidence available to the Administrator to suggest that sources
with flow rates less than 3 MMscf/d or benzene emissions less
than 1 tpy are controlled at the floor, and it was not cost
effective to go beyond the floor.
Comment; Commenter IV-D-10 requested clarification of the
term "benzene emissions to the atmosphere," for the 1 tpy cutoff.
The commenter requested that the term mean actual benzene
emissions.
Response; It was the EPA's intent to specify actual average
benzene emissions and has revised proposed §§63.764(e) and
63.1274{b) (now codified at §63.764(e)(1) of final subpart HH and
§63.1274(d)(1) of final subpart HHH) to clarify that actual
average benzene emissions must be calculated for the 1-tpy
exemption.
Comment: Commenter IV-D-05 stated that, as proposed, the
regulations would exempt a glycol unit that processes less than
3 MMscf/d on an annual average, but is permitted to process more
than 3 MMscf/d annually. The commenter stated that this would
mean that PTE is not a factor as it historically has been in
determining affected units.
Response; The EPA proposed this applicability criteria to
exempt those glycol dehydration units for which the MACT floor
was identified to be no control. These glycol dehydration units
¥
are not exempt from the subpart, but are exempt from the control
requirements of subparts HH and HHH.5 However, records of this
5It should be noted that these criteria are not related to
the determination of PTE. Sources that meet these criteria are
not subject to the control requirements of subparts HH and HHH,
but are still subject the NESHAP. In a previous response, the
EPA has announced the addition of an applicability cutoff for
(continued...)
2-30
-------
actual average throughput level (or the other applicability
criteria) roust be documented and maintained annually to remain
exempted from the control requirements.
The EPA has knowledge of facilities that operate their
glycol dehydration units above their nameplate capacity.
Therefore, maintaining records of design capacity would not
ensure operation below the throughput cutoffs. Therefore,
§§63.774(d)(1) and 63.1284(d)(1) of the final rules specify that
actual annual average natural gas throughput must be maintained,
not the design capacity. Similarly, the final rules contain
criteria for documenting actual average benzene emissions
[codified at §§63.774(d)(2) and 63.1284(d)(2)].
Comment: Commenters IV-D-07 and IV-D-31 requested that the
EPA make provisions in the PTE determination for fluctuations in
water content and gas composition without having to sample the
gas stream frequently.
Response; It is the EPA's understanding that, based on
available information from the production industry, water content
and gas composition remain relatively constant if the source of
the input streams (such as reservoirs) does not change. However,
although dramatic fluctuations in water content and gas
compositions may occur in the transmission and storage industry,
it is believed that they would be over a short period.
Furthermore, since PTE is a worst case calculation, the EPA does
not believe that frequent sampling would be required. However,
sampling would be required if the source of the input stream
5(...continued)
which production facilities and transmission and storage
facilities below this value would not be subject to the NESHAP.
2-31
-------
changed. Therefore, the EPA has not modified subparts HH and HHH
in response to these comments.
Storage Vessels with the Potential for Flash Emissions
Comment: Commenter IV-D-03 requested that the proposed
storage tank exemption/control criteria be based on "credible
engineering test methods supported by fundamental principles of
fluid phase behavior." The commenter provided a Society of
Petroleum Engineers journal article entitled Test Method for
"Actual" True Vapor Pressure of Crude Oils. The article
presented data for flash gas emissions from black oil. According
to the article, a 35° API oil with a GOR of 13.4 scf/bbl had the
flash gas emission potential to exceed benzene, toluene, ethyl
benzene, xylene (BTEX) rates of 10 tpy with less than 5,000
bbl/day. The commenter noted that this oil would have been
exempted from the control requirements. The commenter further
noted that the DOE has degassed this oil to prevent such high
emissions.
Commenter IV-G-01 suggested that breathing, working, and
flashing losses from crude oil storage tanks are "significant"
when storage tanks have gas to oil ratios and API gravities less
than the minimums specified in subpart HH. The commenter
provided an example calculation of tank emissions to show how HAP
emissions from a crude oil storage tank could exceed 5 tons per
year [and exceed 16 tons per year of total reactive organic
compounds (ROC)]. The commenter was concerned that subpart HH,
as currently written, may not reflect the maximum degree of
reduction of HAP emissions for oil and gas production sources.
Response; The criteria of an API gravity equal to or
greater than 40 degrees and an initial producing GOR equal to or
greater than 0.31 m3/liter were used in the proposed rule to
define storage vessels with the potential for flash emissions.
The EPA's analysis of storage vessels that contain hydrocarbon
liquids that have API gravity and initial producing GOR higher
2-32
-------
than these criteria indicate the potential for significant flash
emissions.
The EPA developed the definition for storage vessels with
the potential for flash emissions based on criteria (i.e., API
gravity and GOR) that were easily recognized by industry
personnel and relatively easy to obtain. Furthermore, these
criteria are based on hydrocarbon liquid characteristics.
According to section 112(d)(1), the Administrator is
required to establish emission standards for each category of
major sources. Section 112(d)(1) states that "The Administrator
may distinguish among classes, types, and sizes of sources within
a category or subcategory in establishing such standards. ..."
In addition, section 112(d)(3) states that emission standards for
existing sources in a category may be no less stringent than the
MACT floor.
As stated in a previous response, the EPA has established
that among the class of sources referred to as black oil
facilities, the MACT floor is no control. For the class of
sources defined as storage vessels with the potential for flash
emissions (which includes storage vessels that do not process
black oil), the EPA evaluated "... the average emission
limitation achieved by the best performing 12 percent of the
existing sources (for which the Administrator has emissions
information),. . ." (section 112(d)(3)(A) of the Act). The EPA
determined that the top 12 percent of existing storage vessels
with the potential for flash emissions were controlled.
Comment: Commenter IV-D-03 provided an example, developed
at the Strategic Petroleum Reserve, of a condensate with a GOR
greater than 20,000 standard cubic feet per barrel (scf/bbl) and
a 45° API gravity. The condensate was analyzed using the
2-33
-------
EguiVap™ method. The condensate stream was determined to have
flash gas emissions potential due to true vapor pressures greater
than atmospheric. However, EquiVap™ also identified that after
the liquid was further stabilized by the flash tank, no flash gas
emissions were generated because the liquid had a true vapor
pressure less than atmospheric. According to the commenter, the
"arbitrary exemption/controls criteria would have required costly
recovery and incineration of nonexistent flash gases from this
stream even after it had been properly stabilized by the upstream
flash tank."
Response; The EPA recognizes that there could be specific
situations, such as the ones analyzed by the commenters, where
emissions of an exempted stream are higher than those of a non-
exempted stream. In addition, there are many factors that affect
whether flash emissions occur (e.g., pressure drop between two
tanks, liquid vapor pressure, etc.). However, the EPA believes
that this approach identifies hydrocarbon liquids that have a
potential for significant flash emissions under conditions
representative of industry operations.
Comment: Commenter IV-G-01 requested guidance on how GOR
should be measured.
Response; The final subpart HH requires the owner or
operator to determine the initial producing GOR for the
definition of black oil and stock tank GOR for the definition of
storage vessels with the potential for flash emissions. As
stated in a previous response, this distinction is important
because GOR changes with reservoir pressure. The EPA believes
that requiring a GOR measurement at the stock tank will ensure
that fluids with higher gas content (i.e., a greater potential
for flash emissions) will be subject to the control requirements.
In addition, the EPA has added a definition for initial
producing GOR to subpart HH as follows:
2-34
-------
Initial producing GOR means the producing standard
cubic feet of gas per stock tank barrel at the time
that the reservoir pressure is above the bubblepoint
pressure (or dewpoint pressure for a gas).
The ratio of gas to oil should be constant until the bubblepoint
or dewpoint is reached in the reservoir.6 There are various
methods within the industry available for measuring the GOR but
there is not an approved EPA method. Although the EPA does not
specify the method for subpart HH, the method used by the owner
or operator must achieve a determination of the standard cubic
feet of gas per stock tank barrel (scf/bbl) of the hydrocarbon
liquid.
Comment: Commenter IV-D-01 questioned the basis for
exempting storage vessels with an actual throughput less than 500
BPD [21,000 gallons per day (gal/day)] from control requirements
[§63.764(c) (2)], and requested that the EPA delete the provision.
According to the commenter, 500 BPD is a substantial throughput
for the crude oil production in Louisiana. The commenter stated
that due to the storage vessel exemptions most facilities in
Louisiana would be exempt from this subpart.
Response: The data available to the EPA indicated that for
the class of storage vessels not considered to have the potential
for flash emissions (i.e., with API gravity less than 40° or a
GOR less than 1,750 scf/bbl) and with a hydrocarbon liquid
throughput less than 500 bpd, the MACT floor was no control and
it was not cost effective to go beyond the floor [Air Docket
A-94-04 numbers II-A-01 and II-D-50].
The EPA has added the throughput cutoff criterion to the
storage vessels with the potential for flash emissions definition
in final subpart HH. The final rule states that a storage vessel
Reference 2.
2-35
-------
with the potential for flash emissions is defined as a storage
vessel that contains an actual average hydrocarbon liquid with a
stock tank GOR equal to or greater than 1,750 scf/bbl and an API
gravity equal to or greater than 40 degrees, and a hydrocarbon
liquid throughput equal to or greater than 500 bpd. By adding
the throughput criterion to the definition of storage vessels
with the potential for flash emissions, rather than as a cutoff
specified in proposed §63.764(c)(2), storage vessels that do not
meet the criteria for a storage vessel with the potential for
flash emissions are not considered affected sources in the final
rule and are not included in a facility's potential-to-emit (PTE)
calculation for determining major source status.
Comment: Commenters IV-D-04, IV-D-22, IV-D-34, and IV-D-35
requested that the EPA clarify the averaging period for the
500-BPD exemption criterion for storage tanks. Commenter IV-D-04
assumed that it was meant to be an annual daily average basis.
Commenter IV-D-35 suggested using a five-year rolling average
based on maximum actual tank throughput. Commenter IV-D-34
requested that the storage tank throughput be based on an annual
average. The commenter also suggested that the calculation of
the 500-BPD threshold for storage tanks be based on a method
similar to that proposed for glycol dehydration unit flow rates
in §63.772(b)(1). Commenter IV-D-22 stated that there is no
discussion in subpart HH or the preamble of how to determine
applicability to the 500-BPD threshold for storage vessels. The
commenter recommended that the EPA allow either monitoring of the
flowrate, or other documentation (e.g., sales records) of the
storage vessel flowrate, and that calculation of the 500 BPD
limit be based on an annual average.
Response; As stated in a previous response, the 500-bpd
throughput has been added to the definition of storage vessels
with the potential for flash emissions. Thus, storage vessels
2-36
-------
that do not meet this definition are exempt from subpart HH.
Therefore, the EPA believes that establishing recordkeeping and
reporting requirements for these units would be inappropriate.
However, §63.10(b)(3) contains recordkeeping requirements
for applicability determination. Therefore, owners and operators
with storage vessels that are not subject to subpart HH would be
required under this section to maintain records of the
applicability determination for these storage vessels.
Comment: Conunenter IV-D-07 requested that the EPA clarify
whether the regulation applies to the case where a tank battery
has an average throughput less than 500 bbl/tank but a total
throughput of greater than 500 bbl total.
Response; The throughput applicability criteria for storage
vessels with the potential for flash emissions in final subpart
HH applies to each storage vessel. Thus, a tank battery with a
total actual throughput of more than 500 BPD that consists of
several storage vessels, none of which has an actual average
annual throughput equal to or greater than 500 BPD would not be
subject to subpart HH, provided the GOR and API gravity criteria
are met. Therefore, the EPA has not modified subpart HH in
response to this comment.
Comment: Commenter IV-D-16 recommended that the EPA allow
tanks with a specified percent HAP to be excluded from subpart
HH. The commenter suggested that this would prevent a tank at a
major source, with low HAP contents in the liquid, from being
covered. According to the commenter, controls would not be
effective since HAP emissions would be low due to the low HAP
content in the liquid.
Response; The EPA has established that facilities that
process, store or transfer black oil have a low potential for HAP
emissions and are exempt from control requirements under subpart
2-37
-------
HH. The EPA selected the criteria for defining storage vessels
with the potential for flash emissions using parameters that are
easily determined by the industry. Therefore, the EPA does not
believe that specifying a percent HAP content for hydrocarbon
liquids in subpart HH is necessary. Based on the EPA's knowledge
of the industry, the black oil exemption, by itself, exempts
approximately 85 percent of all tank batteries according to
industry data.
Comment: Commenters IV-D-07 and IV-D-24 stated that they
support exempting storage tanks that have the potential for flash
emissions and a hydrocarbon throughput less than 500 BPD.
However, the commenter IV-D-07 requested an exception for
emergencies. Commenter IV-D-24 stated that the 500-BPD exemption
avoids imposing costly controls on the smallest sources.
Response; Through the startup/shutdown/malfunction
provisions in subpart HH, the EPA has attempted to address those
emergencies that may be encountered by industry. Furthermore,
the throughput exemption is based on an annual average, which
should account for daily fluctuations in throughput. Thus, the
EPA does not believe that an additional exemption is necessary in
subpart HH.
Comment: Commenter IV-G-12 stated that subpart HH is
lacking in that it does not distinguish between flashing and
evaporation. According to the commenter, this lack of
specificity could lead to confusion among sources and regulators
concerning which vessels/substances are covered by the proposed
rule. The commenter suggested that subpart HH be clarified to
specify a temperature/phase relationship or maximum vapor
pressure, as well as specifying the source and HAP content of the
liquid stream that is being stored. The commenter presented
examples. The commenter also stated that as an alternative the
EPA could define the term "flashing" in thermodynamic terms
(i.e., the change of state between liquid and vapor phases that
2-38
-------
is not caused by the addition of thermal energy). The commenter
was interested in exempting produced water from a production
facility as well as lubricating oils, fuels, or other similar
fluids.
Response; Temperature and vapor pressure are very dependant
on stream composition. This variability makes it very difficult
to establish boundary conditions for the types of hydrocarbon
liquids processed in this industry. Furthermore, the EPA
believes that API gravity and 6OR are values that are well
understood by the industry, and are usually readily available.
The EPA does not believe that specifying a percent HAP content or
maximum vapor pressure for hydrocarbon liquids in subpart HH is
necessary. However, to clarify the term flash emissions, the EPA
has added the following language to the definition of storage
vessels with the potential for flash emissions that states "Flash
emissions occur when dissolved hydrocarbons in the fluid evolve
from solution when the fluid pressure is reduced and is not
caused by the addition of thermal energy."
Comment: Commenter IV-G-01 was concerned that HAP and
reactive organic compounds (ROC) emissions from storage vessels
with the potential for flash emissions may be significant with a
throughput less than 500 BPD. The commenter provided an example
calculation for a storage tank with a throughput of 250 BPD,
showing emissions from breathing, working and flashing losses.
The example presented uncontrolled ROC emissions of 4.06 tpy and
controlled ROC emissions of 0.20 tpy. Uncontrolled HAP emissions
(including benzene, hexane, and 2,2,4-trimethylpentane) were
estimated to be 1.31 tpy (uncontrolled) and 0.07 tpy
(controlled).
Response: Based on the EPA's analysis, the storage vessel
applicability cutoffs of hydrocarbon liquid throughput of 500
bpd, a GOR less than 1,750 scf/bbl, and an API gravity less than
2-39
-------
40°, the storage vessels with significant HAP emissions will be
controlled under this regulation. The example provided by the
commenter did not show total HAP emissions greater than the major
source thresholds of 10 tpy for individual HAP (the highest HAP
emissions were estimated to be 0.69 tons of hexane per year,
uncontrolled) or 25 tpy for any combination of HAP (total ROC
emissions were estimated to be 4 tpy, uncontrolled) . Therefore,
since these emissions are well below the major source thresholds,
the EPA maintains that the 500-BPD cutoff is reasonable, and has
not changed the definition of storage vessels with the potential
for flash emissions.
Other Exemptions
Comment: Commenter IV-D-01 questioned the basis for
exempting reciprocating compressors in wet gas service from the
compressor control requirements of §61.242-3, and requested that
the EPA delete the provision. According to the commenter, if
there was a leak, HAP released from a compressor in wet gas
service would be higher than that released from a compressor in
dry gas service, since the concentration of HAP is much higher in
wet gas.
Response; The exemption for reciprocating compressor in wet
gas service is consistent with 40 CFR subpart KKK, the Onshore
Natural Gas Processing Plant New Source Performance Standards
(NSPS). Therefore, the EPA has not removed this exemption from
subpart HH.
2.1.3 Other Applicability Issues
Comment: Commenter IV-D-14 stated that many oil and gas
production facilities are located in remote areas and do not have
a substantive impact on human populations. The commenter asked
what the underlying basis was behind the application of MACT
requirements to HAP sources located in remote areas, whether the
2-40
-------
CAA allowed for remote facilities to be exempted from MACT
standards, and whether the EPA considered this in its rulemaking.
Response: The EPA does not have discretion in setting
standards for major sources of HAP emissions, which must be
implemented nationwide. Therefore, major sources located in
remote areas must still comply with MACT.
Comment: Commenter IV-D-16 was concerned that the
applicability section could be interpreted to mean that refinery
Natural Gas Liquid (NGL) plants could be brought into coverage
since they are at a refinery and at a major source. The
commenter requested that a specific exemption be added to §63.760
for NGL Plants at refineries, to make it clear that it is not the
intent of the regulation to cover refinery NGL Plants. According
to the commenter, the rationale behind this exemption would be
that NGL plants were not considered when the MACT floor was set
and that these plants already have controls put on them by other
regulations (e.g., SIP VOC regulations), therefore no
environmental purpose would be served by drawing them into
subpart HH.
Commenter IV-D-16 was also concerned that the applicability
language could be misinterpreted to mean that existing major
sources that have a single or very few gas wells collocated with
the facility would be included. The commenter explained that a
few wells have been drilled and are producing at existing major
sources. According to the commenter, these plants should not be
subjected to coverage by subpart HH merely because they are major
sources for their primary activities and happen to have a single
or a very few gas or oil wells on-site. The commenter
recommended that the EPA exempt these facilities by making it
clear that subpart HH applies only to oil and natural gas
facilities that are major sources by themselves.
Responset The CAA requires the EPA to regulate major HAP
sources. A major HAP source is defined as "any stationary source
or group of stationary sources located within a contiguous area
2-41
-------
and under common control that emits or has the potential to emit
considering controls. ..." This means that the EPA is
obligated to consider the whole site when determining if a source
is major and to regulate co-located emission sources (e.g.,
production wells), when applicable. It should be noted that
§63.760(d) states that if affected sources (glycol dehydration
units, tank batteries, and ancillary equipment located at natural
gas plants) are not present at a facility, there are no
requirements under subpart HH.
Comment: Commenters IV-D-16 and IV-D-22 were concerned that
proposed §63.760(b) (1) (iii) could be misinterpreted. The
commenters recommended that §63.760(b) should be modified to
clarify that only ancillary equipment located at natural gas
plants are to be considered an affected source. Commenter
IV-D-16 suggested that the EPA's intent was to include ancillary
equipment as an affected source for gas plants in the preamble
(63 FR 6295 and 6304). Commenter IV-D-22 suggested that the
phrase "located at natural gas processing plants" be added to
§§63.760(b) (1) (iii) and (iv) .
Response; To clarify the applicability of subpart HH to
ancillary equipment, the EPA agrees with the commenters that
additional language is necessary, and will add the phrase
"located at natural gas processing plants" to proposed
§63.760(b) (1) (iii) and (iv) [now codified at §§63.760(b) (3) and
(4) of final subpart HH].
Comment: Commenter IV-D-16 stated that the line between
subpart HH and subpart HHH needs to be clarified so the same
sources will not be covered by both rules.
Commenter IV-G-12 stated that they operate gas gathering
systems that accept gas from third party wells at which no
processing or treatment occurs. The commenter explained that
this gas is often gathered and brought to a central compressor
station where it is dehydrated and compressed and transferred to
2-42
-------
a transmission pipeline. Although it seems unlikely that a
regulatory agency would try to aggregate emissions from the
gathering and production operations, the commenter suggested that
situations such as these be clarified in the final subpart HHH.
Commenter IV-D-21 cited two major problems with the EPA's
approach to defining the scope of the source categories. First,
the commenter suggested that subpart HHH lacks a clear definition
that distinguishes natural gas transmission and storage
facilities from natural gas production facilities. According to
the commenter, subpart HH states that the natural gas
transmission and storage source category begins at the point
where natural gas enters the natural gas transmission and storage
source category, but does not define this term. Additionally,
subpart HHH does not define the term "transport or store natural
gas." Therefore, the commenter was concerned that the regulated
entity would be required to draw guidance from the separate
definitions for "natural gas transmission" and "facility"
operating in subpart HHH. However, according to the commenter,
these two definitions provide contradictory guidance. For
example, the commenter interpreted the term "natural gas
transmission" to mean that the transmission and storage source
category would begin only when the natural gas first enters the
pipeline and the source category would not include any
processing, either before or after initial entry into the
pipeline. According to the commenter, this interpretation is
supported by the definition of natural gas transmission by
recognizing that processing can occur in the transmission and
storage source category. The commenter recommended that the
final rule include an express delineation of the source
categories so the regulated community does not have to piece
together the delineation from various provisions throughout
subpart HHH.
According to the commenter, the second problem with subpart
HHH is that it fails to acknowledge that natural gas transmission
and storage facilities commonly process natural gas, both before
and after introduction of natural gas into the main transmission
2-43
-------
line. The commenter cited the following examples of processing
that are integral to natural gas transmission activities by
minimizing the formation of hydrates: dehydration, removal of
CC>2, and extraction of natural gas liquids. The commenter
suggested that this interpretation of the definition of facility
is also inconsistent with the BID, which notes that processing is
included in the transmission and storage source category. The
commenter further noted that processing occurs throughout the
pipeline, the location of which is determined by such factors as
gas quality and geographic location. The commenter requested
that subpart HHH recognize the fact that processing occurs as
part of transmission of natural gas.
To address the apparent lack of delineation between the
production and transmission categories, the commenter recommended
that a distinction between the two categories could be defined by
reference to the point at which that transfer to the transmission
company occurs. The commenter stated that processing operations
that occur prior to the transfer point would fall within the gas
production category, and those performed by the transmission
company after the transfer would fall in the transmission and
storage source category. According to the commenter, the BID
states that natural gas is typically transferred at a meter
station, but may occur at other points. The commenter stated
that dehydrators operated by the transmission company after the
transfer point would fall within the transmission and storage
source category regardless of whether the dehydrators are located
before or along the main transmission line.
Response; The natural gas transmission and storage
definitions in subpart HHH were developed in consultation with
natural gas transmission and storage stakeholders. The EPA
believes that the definitions in subparts HH and HHH delineate
the boundaries of the oil and natural gas production and natural
gas transmission and storage source categories. The key points
in this delineation are (1) the point of custody transfer, which
2-44
-------
is a commonly understood definition within industry, and (2) the
natural gas processing plant/ which is a clearly defined facility
within the production source category. Based on these
discussions with industry, the EPA understood that there was only
one point of custody transfer indicating the point at which
natural gas entered the transmission pipeline. However, the EPA
has made some changes to more clearly define the boundary between
subparts HH and HHH.
The EPA believes that a compressor station located between a
well and a natural gas processing plant or between the well and
the point of custody transfer should be considered part of the
oil and natural gas production source category. Therefore, to
clarify this intent, the final subpart HH states that natural gas
enters the natural gas transmission and storage source category
after the natural gas processing plant, when present
[§63. 760 (a) ] . If no natural gas processing plant is present,
natural gas enters the natural gas transmission and storage
source category after the point of custody transfer. Subpart HHH
also states that compressor stations that transport natural gas
prior to the point of custody transfer, or to a natural gas
processing plant (if present) are considered part of the oil and
natural gas production source category [§63 . 1270 (a) ] , and the
following definition of custody transfer has been added to
§63.1271:
Custody £r^ngf?r means the transfer of hydrocarbon
liquids or natural gas: (1) after processing and/or
treatment in the producing operations, or (2) from
storage vessels or automatic transfer facilities, or
other equipment, including product loading racks, to
pipelines or any other forms of transportation.
2-45
-------
The EPA has also made clarifying changes to the definition
of facility in subparts HH and HHH.
2-46
-------
2.2 DEFINITIONS
2.2.1 Facility
Several commenters responded to the EPA's request for
comments on the definition of facility. The commenters requested
clarification of or suggested changes to the proposed definition
of facility. Commenter IV-D-35 agreed with the EPA's proposed
definition of facility. The following paragraphs present more
detailed comments on the definition of facility.
Comment: Commenter IV-D-02 was concerned that units, which
may include large sections of land and many leases and which are
under the control of a single operator, may be considered a
single facility. According to the commenter, units are created
when groups of leases are combined into a single entity, under
common control, and some States require the formation of units.
The commenter stated that since the terms "site" and "lease" (as
contained in the definition of facility in §63.761) are not
defined in the Act or in subpart HH, it is unclear whether a unit
could be included within the term "lease" and whether a lease
retains it's identity when included in a unit.
Therefore, the commenter requested that the EPA define
facility to mean "the equipment at each individual well site and
each individual tank battery or each individual gas or oil
treating emplacement not located at the wellhead or tank
battery."
Commenters IV-D-05 and IV-D-14 were confused about whether
contiguous surface sites under common ownership would be
considered separate facilities. Commenter IV-D-05 requested that
the EPA modify the definition of facility in subpart HH to
exclude contiguous graded pad sites. Commenter IV-D-14 stated
that the term "surface site" could be interpreted as a single
concrete pad, a grouping of concrete pads in a contiguous area,
or a large graded area without any pads. According to the
commenter, subpart HH could be interpreted to mean that two or
more compressors on adjacent pads in a contiguous area are either
separate facilities or single facilities. The commenter noted
2-47
-------
that this could also apply to several groupings of equipment used
for different purposes but in the same graded area. The
commenter requested that the EPA clarify the definition of
facility to remove these uncertainties.
In addition, commenters IV-D-14 and IV-D-17 stated that
subpart HH does not address the issue of ownership or its effects
on the determination of what makes up a facility. The commenters
asked whether equipment under separate ownership at the same
surface site or adjacent surface sites could be considered a
single facility. Commenter IV-D-17 suggested that, for
clarification, a "facility" be defined to only include equipment
within the boundaries of an individual surface site that operate
under common ownership (e.g., central tank battery, graded pad
site, etc.).
Response; The EPA developed the proposed definition of
facility to: (1) identify criteria that define a grouping of
emission points that meet the intent of the language contained in
section 112 (a) (1) of the Act: "... located within a.
contiguous area and under common control, ..." and (2) contain
terms that are meaningful and easily understood within the
regulated industries. The proposed definition was based on
individual surface sites and the idea that equipment located on
different oil and gas properties (oil and gas lease, mineral fee
tract, subsurface unit area, surface fee tract, or surface lease
tract) shall not be aggregated. In addition, the proposed
definition of a production field facility was limited to glycol
dehydration units and storage vessels with the potential for
flash emissions.
The EPA intended that the facility definition, as it applies
to the oil and natural gas production source category, should
lead to an aggregation of emissions in a major source
2-48
-------
determination that is reasonable, be consistent with the intent
of the Act, and be easily implementable.
The EPA believes that it would not be reasonable to
aggregate emissions from surface sites that are located on the
same lease, but are great distances apart. The definition of
facility states that equipment located on different oil and
natural gas properties (e.g., leases) are not to be aggregated.
Although units (which are made up of more than lease or tract)
are under common control, under the definition of facility, the
equipment located on different leases contained within each unit
would not be aggregated.
Under section 112(a)(1) of the act, a major source is
defined as ". . . any stationary source or group of stationary
sources located within a contiguous area and under common
control. ..." The EPA believes that by defining facility based
on individual surface sites, the EPA has provided relief for
individual surface sites that are located on the same lease, but
are far apart, and excluding contiguous surface sites located on
the same lease would be contrary to the intent of the Act.
Finally, the terms contained in the definition of facility
(e.g., surface site and lease) are well understood within the
industry and by enforcement agencies and the EPA does not believe
that additional definitions or clarifications regarding these
terms are necessary.
Comment: Commenter IV-D-05 suggested adding the term
"permitized area" to the definition of facility in subpart HH.
The commenter stated that mineral leases give operators control
over large tracts of land. According to the commenter, adding
the term "permitized area" would clarify the definition of
facility where production equipment or equipment groupings on
different oil and gas leases are described.
2-49
-------
Response; The EPA believes that adding the term "permitized
area" would add confusion to the definition of facility because
some facilities may not have established "permitized areas," and
different permitting authorities may define permitized areas
differently. The EPA has not made the requested changes to the
definition of facility in subpart HH.
Comment; Commenter IV-D-14 stated that subpart HH does not
address the role of the process in which a grouping of equipment
is engaged, in making the determination of what is a facility.
The commenter explained that a graded area might contain a
compressor station, one or more tank batteries, and a separate
natural gas liquids plant. According to the commenter, each
grouping of equipment may be considered a separate facility by
the owner or operator, and each grouping may be found on separate
concrete pads in a contiguous area under common control. The
commenter asked whether these distinct operations would be
considered a single facility under subpart HH, and requested the
clarification of these issues.
Response; The definition of major source, as proposed in
subpart HH (§63.761), has the same meaning as in §63.2, except
that "emissions from processes, operations, or equipment that are
not part of the same facility, as defined in this section, shall
not be aggregated." A facility, as currently defined in §63.761,
includes equipment within the boundaries of an individual surface
site. The EPA believes that functionally-related equipment is
generally located at the same surface site. Thus, the EPA
believes that any grouping of equipment located on separate
concrete pads (i.e., separate surface sites) would not be
functionally-related; any grouping of equipment on separate
surface sites would be treated as a separate facility for which
emissions would not be aggregated. Furthermore, equipment
located on the same surface site may be a separate facility,
2-50
-------
depending on where the point of custody transfer is within the
facility (e.g., the point at which natural gas enters a natural
gas processing plant is a point of custody transfer and thus the
natural gas processing plant would be considered a separate
facility located on the same surface site).
Comment: Commenter IV-D-16 recommended that the definition
of facility be modified to clarify that compressor engines are
not covered by subpart HH. The commenter noted that vents for
compressor engines are being covered by the Industrial Combustion
Coordinated Rulemaking (ICCR) and should not be covered under
subpart HH.
Response; Facility-wide HAP emissions after the point of
custody transfer must be included in the major source
determination. The EPA does not believe that engine vents from
compressors should be excluded from the definition of facility in
subpart HH because HAP emissions from these units would not be
aggregated for major source determinations.
In addition,, §63.760 (b) (1) specifies the affected sources
for subpart HH. Therefore, the EPA believes that there should be
little confusion, about which emission points are regulated under
subpart HH. Therefore, the EPA has not specifically excluded
engine vents for compressor engines from subpart HH requirements.
Comment: Commenters IV-D-08 and IV-D-22 recommended
clarification to the definition of facility in subpart HH to
include surface units and separate surface sites as tracts on
which multiple groupings of equipment may be located without
those separate groupings being designated as a "facility." In
addition, the commenters recommended that the definition specify
that connection by a road, a waterway, etc. does not cause two
separate groupings of equipment at different sites to be part of
the same facility. The commenters recommended the following
changes:
2-51
-------
Facility means any grouping of equipment (1) where
hydrocarbon liquids are processed, upgraded, or
stored prior to the point of custody transfer, or
(2) where natural gas is processed, upgraded, or
stored prior to entering the natural gas
transmission source category.
For the purpose of a major source determination,
facility (including a building, structure, or
installation) means oil and natural gas production
and processing equipment that is located within
the boundaries of an individual surface site.
Equipment that is part of a facility will
typically be located within close proximity to
other equipment located at the same facility.
Pieces of production equipment or groupings of
equipment located on different oil and gas leases,
mineral fee tracts, lease tracts, subsurface pr
surface unit areas, surface fee tracts, surface
lease tracts, or separate surface sites, whether
or not connected by a road, waterway, walkway.
power line or pipeline, shall not be considered
part of the same facility. Examples of facilities
in the oil and natural gas production source
category include, but are not limited to, well
sites, satellite tank batteries, central tank
batteries, graded pad sites, and natural gas
processing plants.
Response; The EPA agrees with the commenter's
recommendations and has made the suggested changes to the
definition of facility in §63.761, except that the term "walkway"
has not been included in the definition. The EPA believes that
including this term would cause confusion for inspectors because
a walkway between pieces of equipment could become a part of the
boundary of a facility.
Comment: Commenter IV-D-29 recommended that the EPA expand
its definition of production field facility in subpart HH to
include additional HAP emission points beyond glycol dehydration
units and storage vessels with flash emission potential. The
commenter stated that several sources in Santa Barbara and
Ventura Counties that would otherwise be controlled would be
exempt from subpart HH under the proposed definition.
2-52
-------
Response; One of the EPA's objectives was to develop a
definition of facility that would comply with section 112(n)(4)
of the Act and at the same time, reduce the burden on owners and
operators in making a major source determination. The EPA's
evaluation of HAP emission sources in production field operations
suggested that other potential HAP emission points at these
facilities (e.g., equipment leaks) would be inconsequential to
the determination of a facility's major source status. The EPA
believes that eliminating the need to quantify HAP emissions from
small sources at production field facilities, would not affect
the major source status determination, but would reduce the
burden on owners or operators.
Comment: Commenters IV-D-06 and IV-D-31 requested that the
EPA clarify, within the definition of facility in subpart HHH,
whether the EPA intended to exclude facilities used to store
natural gas after the gas enters the local distribution system of
a gas utility. Commenter IV-D-06 interpreted §63.1270(a) to mean
that the affected source runs all the way to the affected end
user, even if some local distribution company exists between the
natural gas transmission and storage source and the end user.
The commenter remarked that according to the preamble, this was
not the EPA's intent. The commenter stated that the affected
source is supposed to run all the way to the end user only if
there is no local distribution company. The commenter
recommended the following language to clarify §63.1270(a):
(a) This subpart applies to owners or operators
of natural gas transmission and storage facilities that
transport or store natural gas prior to entering the
pipeline to a local distribution company or (if therg
is no local distribution company) to a final end user,
and that are major sources of hazardous.air pollutant
(HAP) emissions.
The commenter also recommended changes to the definition of
facility to clarify this point.
2-53
-------
Facility means any grouping of equipment where
natural gas is processed, compressed, or stored prior
to entering a pipeline to a local distribution company
or (if there is no local distribution company) to a
final end user. A facility for this source category
typically is: A natural gas compressor station that
receives natural gas via pipeline, from an underground
natural gas storage operation, from a condensate tank
battery, or from a natural gas processing plant; or An
underground natural gas storage operation. The
emission points associated with these phases include,
but are not limited to, process vents. Processes that
may have vents included, but are not limited to,
dehydration, and compressor station engines. Facility,
for the purpose of a major source determination, means
natural gas transmission and storage equipment that is
located inside the boundaries of an individual surface
site and is connected by ancillary equipment, such as
gas flow lines-;—roadsr or power lines.7 Equipment that
is part of a facility will typically be located within
close proximity to other equipment located at the same
facility. Natural gas transmission and storage
equipment or groupings of equipment located on
different gas leases, mineral fee tracts, lease tracts,
subsurface unit areas, surface fee tracts, or surface
lease tracts shall not be considered part of the same
facility.
The commenter stated that this comment may also apply to subpart
HH.
Response; The affected source in the natural gas
transmission and storage source category should run all the way
to the end user only if there is no local distribution company.
Therefore, the EPA has added the phrase "if there is no local
distribution company" to §63.1270(a) and the definition of
facility in subpart HHH. The EPA also agrees that roads are not
equipment and has removed the term from the definition of
facility in §63.1271 of subpart HHH.
Comment: Commenters IV-D-07 and IV-D-31 stated that since
natural gas storage takes place in depleted gas wells, and
liquids are transferred for processing to the plant, the
7This particular change is for an unrelated reason: roads
are not "equipment."
2-54
-------
definition of facility suggests that a natural gas storage
facility could qualify as a production facility. The commenters
stated that these terms must be clarified to avoid this
misunderstanding.
Responset Subpart HH contains a definition of field natural
gas which means " . . .natural gas that is extracted from a
production well prior to entering the first stage of processing,
such as dehydration." A production well is defined in §63.761 as
a "... hole drilled in the earth from which ... field natural
gas is extracted." Since the gas handled by a natural gas
storage facility has been dehydrated, the natural gas handled by
a storage facility would not be considered field natural gas.
Therefore, given the definitions of production well and field
natural gas, a natural gas storage field that uses a depleted gas
well would not qualify as a production facility. The EPA does
not believe that clarification to the definition of facility is
necessary in response to this comment.
Comment: Commenter IV-D-07 stated that in the preamble, the
term "upgraded" is used concerning hydrocarbon liquids, but is
not defined and should not be included in subpart HH.
Response; The EPA agrees that clarification to the
definition of the term "upgraded" is necessary. Therefore, the
EPA has modified the definition of facility in subpart HH to
specify that "upgraded" means "the removal of impurities or
other constituents to meet contract specifications."
Comment: Commenter IV-D-14 requested that the EPA clarify
the term graded pad in subpart HH. According to the commenter,
clarification of this term is critical in establishing the limits
of a given "facility."
Response; The term graded pad is a term that is commonly
used in the industry. However, this term is used in the
2-55
-------
definition of surface site, and is not appropriate as an example
of a facility. Therefore, £he EPA has removed this term from the
definition of facility.
Comment : In their supplemental comments (IV-G-23},
commenter IV-D-22 recommended that the EPA modify the definition
of facility to clarify that product loading rack equipment falls
within the EPA's definition of facility. The commenter
recommended the following changes:
Facility means any grouping of equipment -f±-)- where
hydrocarbon liquids, natural gas or natural gas liquids
are processed, upgraded, or stored prior to a point of
custody transfer^ — or — (-2-) — where natural gas is
L oo£S5£:Ci , upgraded, oir s t o IT £ ci ]^LXOL Co £nc£irxii Li~i£
ri&uux~d. -L g&s t. ird.ns[i\_L ss ion &oux~c£ c&t«£cjox"y .
Response; The EPA does not believe that the continent er ' s
recommendations are necessary. Natural gas liquids are defined
in §63.761 as hydrocarbon liquids. Therefore, they are already
covered in the definition of facility in subpart HH.
2.2.2 Other Comments on Definitions
Affected Source
Comment : Commenters IV-D-08 and IV-D-22 recommended the
following new definition for affected source, which is "important
to the successful implementation of subpart HH" :
Affected source: For major sources, each emission
point located at a facility that meets the criteria
specified in paragraph (a) and listed in paragraphs
(b) (1) (i) through (b) (1) (iv) of Section 63.760; for
area sources, each TEG dehydration unit located at a
facility that meets the criteria specified in paragraph
(a) of Section 63.760.
Response ; The EPA believes that the addition of the term
affected source to §63.761 is unnecessary since it is defined in
§63.760(b) .
Ancillary Equipment
Comment : Commenter IV-D-06 pointed out that subpart HH
provides no definition for the term product accumulator vessel .
The commenter stated that they did not know, and did not think
2-56
-------
the EPA knew, what a product accumulator vessel was. According
to the commenter, new terms from the hazardous organic NESHAP
(HON), "surge control vessel" and "bottoms receiver" have
definitions the commenter can understand. The commenter
recommended that the EPA use one or both terms in place of
product accumulator vessel, if appropriate, and that the term
product accumulator vessel be eliminated.
Response; The EPA agrees with the commenter that
clarification to the definition of ancillary equipment is
necessary and has modified the definition as follows:
Ancillary equipment means any of the following
pieces of equipment: pumps, compressors/—pressure
relief devices, sampling connection systems, open-ended
valves, or lines, valves, flanges, and other
connectorsT—or product accumulator vessels.
The term "compressors" was removed because they are listed
separately in the subpart HH.
Comment: Commenter IV-D-22 recommended the following
modification to the definition of ancillary equipment to clarify
that such equipment is subject to subpart HH only if it is
located at a natural gas processing plant:
Ancillary equipment means any of the following
pieces of equipment located at a natural gas processing
plant: pumps, compressors, pressure relief devices,
sampling connection systems, open-ended valves or
lines, valves, flanges and other connectors, or product
accumulator vessels.
Response; Section 63.769(a) states that the equipment leak
standards apply to ancillary equipment at natural gas processing
plants. The EPA believes that specifying that ancillary
equipment is located at natural gas processing plants within the
ancillary equipment definition would be redundant. The EPA has
not altered the definition of ancillary equipment in response to
this comment.
2-57
-------
Black Oil
Comment: Commenter IV-D-22 recommended the following
modification to the definition of black oil in subpart HH to
clarify the point of measurement and averaging period for the
GOR:
Black oil means hydrocarbon (petroleum) liquid
with an annual average wellhead gas-to-oil ratio (GOR)
less than 50 cubic meters (1,750 cubic feet) per barrel
and an API gravity of less than 40 degrees for the
stgrage tank liquids.
Commenter IV-G-01 stated that the proposed definition of
black oil does not state where the GOR applies and should be
clarified. The commenter stated that it was not clear if the GOR
applies at the storage tank where flashing occurs or at the
subsurface reservoir.
Response; The EPA intends that the GOR should be measured
as the initial producing GOR, rather than the average wellhead
GOR. The EPA has added the phrase "initial producing" before GOR
to the definition of Jblac.k oil. The EPA has also added the
following definition for initial producing GOR to §63.761.
Initial producing GOR means the producing standard
cubic feet (scf) of gas per stock tank barrel (bbl) at
the time that the reservoir pressure is above the
bubblepoint pressure (or dewpoint pressure for a gas).
Boiler
Comment: Commenter IV-D-06 recommended that the EPA amend
the proposed definition of boiler to include resource
conservation and recovery act (RCRA) industrial furnaces. The
commenter explained that this change is necessary to provide
those devices an exemption from performance testing. The
commenter stated that the EPA should use the definition of boiler
in §63.111 of subpart G of the HON, verbatim, as the definition
of boiler in subpart HHH. The commenter stated that this comment
also applies to subpart HH. The commenter suggested that an
alternative would be to add separate provisions for industrial
2-58
-------
furnaces; however, the cotnmenter noted that this would require.
more extensive drafting. Since all the relevant cross-references
to RCRA regulations are the same for industrial furnaces as for
boilers and the commenter stated that adding industrial furnaces
to the definition of a boiler would be easier.
Commenter IV-D-22 recommended the following modification to
the definition of boiler to be consistent with the ICCR:
Boiler means an enclosed device using controlled
flame combustion and having the primary purpose of
recovering and exporting thermal energy in the form of
steam or hot water.
Response; The EPA is not aware of any oil and natural gas
production or natural gas transmission and storage facilities
that would have RCRA industrial furnaces. However, the EPA does
not see any reason to not incorporate the commenter's (IV-D-06)
suggested language. In addition, the EPA has modified the
definition of boiler in subparts HH and HHH to be consistent with
the ICCR. The following definition of boiler has been added to
§§63.761 and 63.1271:
Boiler means an enclosed device using controlled
flame combustion and having the primary purpose of
recovering and exporting thermal energy in the form of
steam or hot water. Boiler also means any industrial
furnace as defined in 40 CFR 260.10.
Closed Vent System
Comment: Commenter IV-D-06 requested that the EPA clarify
that closed-vent systems vent emissions to control devices and
not to a process. The commenter explained that process piping
routes emissions to (or from or within) a process. The commenter
further explained that process piping may have equipment subject
to equipment leak monitoring requirements that should not be
subject to closed-vent system monitoring requirements.
Therefore, the commenter suggested that the EPA make the
following changes to subparts HH and HHH:
2-59
-------
(1) Revise the definition of closed-vent system in §63.1271 as
shown :
Closed-vent system means a system that is not open
to the atmosphere and is composed of piping, ductwork,
connections, and if necessary, flow inducing devices
that transport gas or vapor from an emission point to a
control device or back into the process . If gas or
vapor from regulated equipment is routed to a process
(e.g., to a fuel gas system), the process conveyance
system shall not be considered a closed vent system and
is not subject to closed vent system standards.
(2) Revise the definition of control device as shown:
Control device means any equipment used for
recovering or oxidizing hazardous air pollutant (HAP)
and volatile organic compound — (VOC) 8 vapors . Such
equipment includes, but is not limited to, absorbers,
carbon adsorbers, condensers, incinerators, flares,
boilers, and process heaters. For the purposes of this
subpart, if gas or vapor from regulated equipment is
used, reused, returned back to the process, or sold,
then the recovery system used, including piping,
connections, and flow inducing devices, is not
considered to be control devices or closed-vent
systems .
(3) Revise §63 . 1275 (c) (1) as shown:
(1) The owner or operator shall control air
emissions by connecting the process vent to a process
natural gas line through — a — closed" veil L — system designed
o/£L&t~£iJ. .tri &ocord&nc^ wit.ri tri£ ^. £m x ^mti'iti OTT
section 63. 1281 (c) — and — (tit".
Response; The definitions of closed vent system and control
device and §63 . 1275 (c) (1) of subpart HHH, and §63 . 765 (c) (1) of
subpart HH, have been revised as follows, to clarify that
8This specific change is unrelated, but important. Subpart
HHH is not a VOC rule; it is a HAP rule. Additionally, the way
the proposed definition was worded, nothing could be a control
device unless it controlled both HAP and VOC. In other words, if
a device to control only HAP was installed, that device could not
be a control device because it is not also controlling VOC. Such
a result would be nonsensical, but it would appear to be
compelled by the literal wording of the proposed definition. By
eliminating the words "and volatile organic compound (VOC)" from
the definition, the EPA can resolve that difficulty.
2-60
-------
closed-vent systems vent emissions to control devices and not to
a process:
Closed-vent system means a system that is not open
to the atmosphere and is composed of piping, ductwork,
connections, and if necessary, flow inducing devices
that transport gas or vapor from an emission point to a
one or more control devices, or back into tho process.
If gas or vapor from regulated equipment is routed to a
process (e.g., to a fuel gas system), the process
conveyance system shall not be considered a closed-vent
system and is not subject to closed-vent system
standards.
Control device means any equipment used for
recovering or oxidizing hazardous air pollutant
(HAP) or asd-volatile organic compound (VOC)
vapors. Such equipment includes, but is not
limited to, absorbers, carbon adsorbers,
condensers, incinerators, flares, boilers, and
process heaters. For the purposes of this
subpart, if gas or vapor from regulated equipment
is used, reused (i.e.. iniected into the flame
zone of a combustion device). returned back to the
process, or sold, then the recovery system used,
including piping, connections, and flow inducing
devices, is not considered to be control devices
or closed-vent systems.
Regarding the removal of "volatile organic compounds" from
the definition of control device, the EPA does not believe that
including VOC suggests that the Agency is regulating VOC.
Compressor Station
Comment: Commenter IV-D-06 stated that the phrase "supplies
energy" in the definition of compressor station is vague.
According to the commenter, under a literal interpretation of
subpart HHH, a power plant would be a "compressor station"
because it "supplies energy" to run the compressors that move the
gas. The commenter contended that the EPA intended that
compressor stations would have compressors. The commenter
suggested the following language for clarification:
2-61
-------
Compressor station means any permanent combination
of equipment compressors that suppliesenergy to move
natural gas at increased pressure from fields, in
transmission pipelines, or into storage.
The commenter stated that this comment may also apply to
subpart HH.
Response; In order to clarify the Agency's intent/ the EPA
has modified definition of compressor station in subpart HHH as
suggested by the commenter.
Condensate
Comment: Commenter IV-D-22 recommended the following
modification to the definition of condensate in subpart HH to
specify what the standard conditions are:
Condensate means hydrocarbon liquid that condenses
because of changes in temperature, pressure, or both,
and remains liquid at standard conditions of 14.7
pounds per square inch, absolute (psia) and 60°F.
Commenter IV-D-02 recommended that the EPA define condensate
as MMS does at 30 CFR §206.51 as follows:
Condensate is a mixture of liquid hydrocarbons
that result from condensation of petroleum hydrocarbons
existing initially in a gaseous phase in the
underground reservoir.
Response; The EPA agrees that the definition of condensate
needs to refer to liquids produced from natural gas. In
addition, the definition of standard conditions (68 °F and 29.92
in Hg) is provided in subpart A (§63.2). Therefore, the revised
definition of condensate is as follows:
Condensate means hydrocarbon liquid separated from
natural eras that condenses due to because of changes in
the temperature, pressure, or both, and remains liquid
at standard conditions, as specified in S63.2 of this
part.
Condenser
Comment: Commenter IV-D-05 recommended that the EPA add a
definition for the term condenser. The commenter stated that
many still column vents have been modified to add tubing to the
2-62
-------
normal exhaust port. According to the commenter, if an exhaust
experiences a temperature decrease because of a modification,
then any amount of tubing or apparatus added to decrease the
temperature could be classified as a condenser.
Response: The EPA does not agree that a definition of
condenser is necessary. It was not the EPA's intent to allow the
kind of scenario described by the commenter as a control
technology.
Continuous Seal
Comment: Commenter IV-D-06 interpreted the definition of
continuous seal to require a single-piece seal. According to the
commenter, some seals, consisting of more than one piece, make a
continuous seal. The commenter recommended that the EPA use
language similar to that used in the HON, which allows for seals
consisting of more than a single piece, for subpart HH.
Response; The term continuous seal is not used in subpart HH
and has been removed from §63.761. In addition/ the term fill or
filling is not used in subpart HH and has also been deleted.
Custody Transfer
Comment: Commenter IV-D-05 recommended that the term
custody transfer in subpart HH be clarified to account for the
case where a gas processing plant is incorporated within an oil
and gas production facility to process the gas further. The
commenter explained that in such cases, the gas does not leave
the pad site and does not change ownership.
Response; The EPA considers the point at which natural gas
enters a natural gas processing plant as a point of custody
transfer. Therefore, a natural gas processing plant is a
facility, despite whether or not the processing plant is
incorporated within the production facility. The EPA has not
made any changes to subpart HH in response to this comment.
Comment: Commenter IV-D-16 stated that the definition of
custody transfer in subpart HH is not consistent with the
2-63
-------
definition of custody transfer in other rules. [Note: the
commenter did not provide examples of the other rules.] The
commenter recommended that the phrase "for this regulation" needs
to be added to clarify that this definition is only for subpart
HH, so no misunderstandings will occur.
Response; The EPA has not made the commenter * a recommended
changes to the definition of custody transfer. The definition of
custody transfer was derived from 40 CFR part 60, subpart Kb.
Definitions in other rules would apply to those rules regardless
of how the EPA defines custody transfer in subpart HH, therefore
the commenter's clarification is unnecessary.
Comment: Commenter IV-D-16 recommended that the definition
of custody transfer needs to address that a custody transfer
still occurs when a subsidiary or different branch of the same
company "sells" or transfers natural gas to another branch of the
company.
Response; According to the proposed definition of custody
transfer, the scenario described by the commenter would be
considered a custody transfer, if gas was transferred from
processing and/or treatment in the producing operations, from
storage vessels or automatic transfer facilities, to pipelines or
any other forms of transportation or if the gas was transported
to a natural gas processing plant. The scenario would not be
considered a point of custody transfer if the gas only changes
ownership within the company. Therefore, the EPA has not made
any changes in response to this comment. The EPA intends to
issue implementation guidance on applicability in the future.
Comment: According to commenter IV-D-22, the definition of
custody transfer in subpart HH does not need to include natural
gas since applicability for natural gas production is established
by the gas entry into the facility subject to the natural gas
transmission and storage source category. Therefore, the
2-64
-------
commenter recommended the following modification to the
definition of custody transfer to refer only to hydrocarbon
liquids:
Custody transfer means, for purposes of this
subpart, the transfer of hydrocarbon liquids after
processing and/or treatment in the producing
operations, from storage vessels or automatic transfer
facilities to pipelines or any other forms of
transportation.
However, in their supplemental comments (IV-G-23), commenter
IV-D-22 requested that the EPA clarify that the point of custody
transfer is beyond such equipment as product loading racks so
that this equipment is covered by subpart HH. The commenter
stated that their position presented in their original comment
letter (i.e., to remove natural gas from the definition of
custody transfer) had changed. The commenter stated that they
believed that it would be prudent for the EPA to adopt the
following definition of custody transfer:
Custody transfer means the transfer of hydrocarbon
liquids or natural gas, after processing and/or
treatment in the production operations, from storage
vessels, automatic transfer facilities, or other such
equipment, including product loading racks, to
pipelines or any other forms of transportation. For
the purposes of this subpart. the EPA considers the
point at which such liquids or natural gas are placed
into pipelines or other forms of transportation to be a
point of custody transfer.
Response; The EPA agrees with the commenter. If the term
natural gas were removed from the definition of custody transfer,
the definition of associated equipment would be extended to the
end of the natural gas processing plant. During discussions with
industry, the EPA believes that it was clear that aggregating
emission points from natural gas processing plants was an
acceptable interpretation of section 112(n)(4). The EPA has also
made the recommended change to the definition of custody transfer
to incorporate loading rack equipment.
EcmipTnent. Leaks
2-65
-------
Comment: Commenter IV-D-06 stated that the definition of
equipment leak is inconsistent with the definition of ancillary
equipment. For example, the commenter stated that product
accumulator vessels are not included under the definition of
equipment leak but they are included under the definition of
ancillary equipment. The commenter noted that other differences
between the two definitions are possible. According to the
commenter, §63.769 (equipment leak standards) applies to
"ancillary equipment" however, "that does not seem to work,"
especially for "ancillary equipment" that is not defined as
"equipment leaks." The commenter suggested that subpart HH
"could get by with" either a definition of "equipment leak" or a
definition of "ancillary equipment."
Response; The EPA has modified the definition of equipment
leaks as follows to remove inconsistencies between the
definitions of equipment leak and ancillary equipment:
Equipment leaks means emissions of hazardous air
pollutants from ancillary equipment (as defined in this
section) and compressors.
Federally enforceable
Comment: Commenter IV-D-05 stated that the definition of
major source allows a unit to consider control devices when
making major source determinations. According to the commenter,
the preamble addresses this and makes it clear that the controls
must be federally enforceable. The commenter assumed that
federally enforceable would apply if it were incorporated as a
condition of an operating permit that has gone through the
title V process (including the public comment period) . If this
is not the correct interpretation, the commenter stated that the
term federally enforceable was vague and should be clarified.
Response; The commenter's interpretation was correct. The
definition of the term federally enforceable, in §63.2 of the
General Provisions, includes a list of federally enforceable
2-66
-------
terms and conditions, which includes those contained in a title V
permit. The EPA believes that §63.2 is clear about what is
considered federally enforceable. Therefore, the EPA has not
incorporated a definition of federally enforceable that is
specific to subparts HH and HHH.
Field Natural Gas
Comment: Commenter IV-D-22 recommended deleting the
definition of field natural gas from subpart HH. The commenter
stated that the term was not sufficiently different from "natural
gas" to require a separate listing.
Response; The term field natural gas is necessary to
distinguish between a natural gas production well and a natural
gas storage facility that uses a depleted well for storage. A
natural gas production well extracts natural gas that has not
been processed (i.e., field natural gas) and a natural gas
storage facility extracts natural gas that has been processed.
Therefore, the EPA has not removed the definition of field
natural gas from §63.761.
Flow Indicator
Comment: Commenter IV-D-06 recommended that the EPA clarify
that a flow indicator can include a device that shows the
position of a valve, rather than necessarily requiring a direct
reading of "flow." The commenter stated that some EPA inspectors
have said that a valve position indicator is not a flow indicator
because it does not directly detect "flow." The commenter
suggested that the literal language of some rules (before the
HON) might support this position. The commenter suggested that
the EPA use the current, amended definition of flow indicator
from §63.111 of subpart G of the HON and use it in subpart HHH,
verbatim. The commenter stated that this comment also applies to
subpart HH.
2-67
-------
Response; The EPA agrees that a flow indicator should show
valve position rather than directly reading the flow. Therefore,
the EPA has included the following definition of flow indicator
in final subparts HH and HHH:
Flow indicator means a device which that indicates
whether gas flow is present in a line or whether the
valve position would allow gas flow to be present in a
line.
Glycol Dehydration Unit
Comment: Commenter IV-D-06 stated that the definition of
glycol dehydration unit in subpart HHH has loopholes. According
to the commenter, a unit could be reconfigured so the natural gas
was not running counter currently to the glycol stream, meaning
the unit would not be considered a "glycol dehydration unit" and
would not be regulated. The commenter further explained that a
unit would not be considered a "glycol dehydration unit" and
would not be regulated if it were reconfigured to (1) regenerate
glycol by any method other than distillation, or (2) not recycle
glycol back to "the absorber." The commenter provided the
following language to address these loopholes.
Glycol dehydration unit means a device in which a
liquid glycol directly contacts a natural gas stream
\t,IiciL. IS w*nrGU._L"cTtGCi C-wU.li^"d!r dU.J-J.'ISllL. tO U..T1C.: yJ.^'CG-1- H -L O W /
and absorbs water in a contact tower or absorption
column (absorber). The glycol contacts and absorbs
water vapor and other gas stream constituents from the
natural gas and becomes "rich" glycol. This glycol is
then regenerated by—distilling the—wateu—and—otliei"—gas
stream coristitueulb in the glycol dehydration unit
reboiler. The distilled or "lean" glycol is then
recycled back to the absorber.
The commenter stated that this comment may also apply to subpart
HH.
Commenter IV-D-22 recommended the following addition to the
definition of glycol dehydration unit to clarify the types of
units covered by subpart HH:
Glycol dehydration unit means a device in which
liquid glycol (ethylene glycol. diethylene glycol. or
2-68
-------
triethylene glycol) absorbent directly contacts a
natural gas stream (that is circulated counter current
to the glycol flow) and absorbs water vapor in a
contact tower or absorption column (absorber)....
Response; The EPA agrees that the definition of glycol
dehydration unit in subparts HH and HHH has loopholes.
Therefore, the EPA has modified the definition to remove any
possible loopholes and to add some examples of types of glycol
that may be used in the process. Final subparts HH and HHH
contain the following revised definition of glycol dehydration
unit:
Glycol dehydration unit means a device in which a
glycol. diethylene glycol. or triethylene glycol)
absorbent directly contacts a natural gas stream (that
is circulated counter current to feho glyeol flow)—and
absorbs water vapor in a contact tower or absorption
column (absorber). The glycol contacts and absorbs
water vapor and other gas stream constituents from the
natural gas and becomes "rich" glycol. This glycol is
then regenerated by distilling water and other gao
stream constituents in the glycol dehydration unit
reboiler. The distilled or "lean" glycol is then
recycled back to the absorber.
Hydrocarbon Liquid
Comment: Commenter IV-D-22 recommended the following
modification to the definition of hydrocarbon liquid to clarify
that produced water is not a hydrocarbon liquid:
Hydrocarbon liquid means any naturally occurring,
unrefined, petroleum liquid; produced water is not a
hydrocarbon liquid.
Response; The EPA has not modified the definition of
hydrocarbon liquid in response to this comment. The statement
that produced water is not a hydrocarbon liquid is contained in
the definition of produced water. The EPA believes that it would
be redundant to include it in the definition of hydrocarbon
liquid.
2-69
-------
Hydrocarbon Throughput
Comment: Commenter IV-G-01 stated that hydrocarbon
throughput should be defined [for the storage tank applicability
criteria of 500 barrels per day (BPD)]. According to the
commenter, naturally occurring hydrocarbon consists of oil,
water, and gas.
Response; The EPA does not believe there is a need to add a
definition for hydrocarbon throughput. The storage tank
applicability criteria in §63.764(c)(2) is based on the "actual
throughput of hydrocarbon liquids." Furthermore, the term
hydrocarbon liquids is defined in §63.761.
In Volatile Organic HAP (VOHAP) Service
Comment: Coinmenters IV-D-06, IV-D-20, and IV-D-22 requested
clarification for the averaging period for the VOHAP
concentration trigger, for a stream to be subject to the
equipment leak standards. Commenter IV-D-06 asked whether the
threshold concentration of 10 percent organic HAP in the
definition of in VOHAP service in subpart HH means annual average
concentration, the normal concentration under standard operating
conditions, or the highest concentration ever encountered. The
commenter stated that if the 10 percent figure applies to the
highest concentration, it will make applicability and enforcement
difficult because the highest concentration will probably not be
present during inspections. Commenters IV-D-06, IV-D-20, and
IV-D-22 recommended an annual average, such as that in subpart H
of the HON or 40 CFR part 63, subpart CC.
Response; The MACT floor for equipment leaks at natural gas
processing plants was determined to be at the level of control
required under the onshore natural gas processing plants NSPS (40
CFR part 60, subpart KKK). The control requirements of 40 CFR
part 60, subpart KKK are equivalent to those in 40 CFR part 61,
subpart V. Since subpart V is a HAP rule, the oil and natural
gas production NESHAP cross references subpart V. The
2-70
-------
requirements in subpart V state that, for a piece of equipment to
be considered not in volatile HAP (VHAP) service, it must be
determined that the percent VHAP can be expected never to exceed
10 percent by weight. Therefore, the EPA has not modified the
averaging period for determination for in VOHAP service.
However, to be consistent with subpart V, and to avoid
confusion between the two rules, the EPA has modified this
definition to refer to VHAP, rather than VOHAP. This change will
also affect several sections within subpart HH, including, the
definition of VOHAP (§63.761), the equipment leaks standards
(§63.769), and test methods and procedures (§63.772).
In Wet Gas Service
Comment: Commenter IV-D-01 stated that in wet gas service
is not defined in §§63.761 or 61.241. The commenter also stated
that it is not defined in either the NSPS or the hazardous
organic NESHAP (HON) equipment leak provisions.
Response; The EPA agrees that a definition of in wet gas
service is necessary. Therefore, the EPA has added the following
definition of in wet gas service to §63.761, based on the one
contained in subpart KKK:
In wet gas service means that a piece of equipment
contains or contacts the field gas before the
extraction of natural gas liquids.
Incinerator
Comment: Commenter IV-D-06 noted that the definition of
incinerator contains the wrong word. The commenter stated that
the last sentence mentions energy recovery sections that "permit"
the incoming vent stream or combustion air and it should say
"preheat." The commenter stated that this comment may also apply
to subpart HH.
Response; The definition of incinerator in subpart HHH has
been revised to chajige the word "permit" to "preheat." A
2-71
-------
definition of incinerator was not included in subpart HH.
Therefore, the same revised definition has been added to §63.761
of final subpart HH.
Natural Gas
Comment: Commenters IV-D-07 and IV-D-31 requested that the
EPA revise the definition of natural gas in §63.1271 to state
that the primary constituent of natural gas is methane, without
reference to any other components. Additionally, according to
the commenters, water vapor is essentially removed before
transmission. The commenters stated that it appears the
definition of natural gas used is more appropriate for production
gas than transmission and storage gas.
Response; The EPA agrees that the definition of natural gas
is confusing. Therefore, the EPA has revised the definition of
natural gas in §§63.761 and 63.1271 based on the definition
contained in the Onshore Natural Gas Processing NSPS, 40 CFR part
60, subpart LLL (§60.641):
Natural gas means a naturally occurring mixture of
hydrocarbon and nonhydrocarbon gases found in geologic
formations beneath the earth's surface. The principal
hydrocarbon constituent is methane.
Natural Gas Liquids
Comment: Commenter IV-D-22 recommended the following
modification to the definition of natural gas liquids in subpart
HH to distinguish liquid phase hydrocarbons from vapor phase
hydrocarbons:
Natural gas liquids (NGLs) means the liquid
hydrocarbons, such as ethane, propane, butane, pentane,
natural gasoline, and condensate that are extracted
from field gas.
Response; The EPA agrees with the coramenter and has
modified the definition of natural gas liquids, as follows:
Natural gas liquids (NGLs) means the liquid
hydrocarbons, such as ethane, propane, butane, pentane.
2-72
-------
natural gasoline, and condensate that are extracted
from field natural gas.
Natural Gas Processing Plant
Comment: Commenter IV-D-07 noted that the definition of a
natural gas processing plant in subpart HH is inconsistent with
the background information document (BID)9. According to the
commenter, "extraction" not "separation and fractionation" is
described as occurring at a natural gas processing plant. The
commenter stated that a definition of "extraction" or
"separation" should be included in the definition. As an
alternative, the commenter recommended that the EPA use the BID
description. The commenter also stated that the definitions of
"field natural gas" and "production well" are inadequate.
Commenter IV-D-11 requested that the EPA clarify the term
"extraction" as it is used in the definition of natural gas
processing plant in subpart HH. The commenter was concerned that
the term "extraction" could be misinterpreted to include simple
producing field separation of natural gas and liquids that occurs
absent of "processing." The commenter stated that the EPA's
discussion of its interpretation of the term "extraction" in the
proposal and background documents for the NSPS for Natural Gas
Processing Plants (subpart KKK) were a correct interpretation for
those activities that occur at a natural gas plant versus those
that occur at field production facilities. The commenter was
concerned that the State employees that interpret the rules for
compliance purposes are not always familiar with natural gas
processes. The commenter recommended that the EPA include a
definition of "extraction" similar to that in the NSPS preamble
in the final subpart HH. The commenter stated that if the EPA
does not include this definition in the final rule, the EPA
9 National Emission Standards for Hazardous Air Pollutants
for Source Categories: Oil and Natural Gas Production and
Natural Gas Transmission and Storage - Background Information for
Proposed Standards. U.S. Environmental Protection Agency.
Research Triangle Park, NC. Publication No. EPA-453/R-94-079a
April 1997.
2-73
-------
should clarify the intent of the term in the response to comments
in the final rule's preamble.
Response t The EPA does not believe that further
clarification is necessary. The EPA has had extensive discussions
with industry and trade associations during the development of
subpart HH related to the definitions of field natural gas and
production well, and developed these definitions based on this
information. Furthermore, the definition of natural gas
processing plant in §63.761 corresponds to the definition in
subpart KKK.
Comment: Commenter IV-D-16 suggested that tighter language
or the addition of exclusionary language should be added to the
definition of natural gas processing plants to clarify that
subpart HH does not cover natural gas liquid (NGL) plants located
at refineries. According to the commenter, as defined, NGL
plants at refineries which take liquid NGLs into the plant and
fractionate them into pure NGL streams would be included.
Response; It was not the EPA's intent to regulate NGL
plants at refineries under the provisions of subpart HH.
However, the it was the EPA's intent that natural gas processing
plant mean any processing site engaged in both of the criteria
listed in the proposed definition. Therefore, the EPA has
modified the definition of natural gas processing plant as
follows:
Natural gas processing plant (gas plant) means any
processing site engaged in-;—Rrf the extraction of
natural gas liquids from field gas, or 4J9—the
fractionation of mixed NGLs to natural gas products, or
a combination of both.
Comment: Commenter IV-D-22 recommended the following
modification to the definition of natural gas processing plant in
subpart HH to identify the gas plant according to its primary
2-74
-------
activities and Standard Industrial Classification (SIC) or North
American Industry Classification System (NAICS) code:
Natural gas processing plant (gas plant) means any
processing site, engaged in the primary purpose of
which is (1) the extraction of natural gas liquids from
field gas, (2) the fractionation of natural gas liquids
to natural gas products, or both, and which is
classified in SIC Code 1321 (NAICS Code 211112) . For
purposes of subpart HH. a aas plant is considered a
facility. A major source determination for a natural
gas processing plant will aggregate HAP emissions from
the facility, between the inlet scrubber and the plant
tailgate or other facility outlet boundary.
Response ; The EPA has not made the recommended modification
to the definition of natural gas processing plant. The EPA does
not believe that including an SIC code in the definition of
natural gas processing plant is necessary. Furthermore, the EPA
does not believe that the commenters last two suggested sentences
(beginning with "For the purposes of subpart HH . . . n) are
necessary because the intent contained within these sentences is
included in the definition of facility.
No Detectable Emissions
Comment : Comtnenter IV-D-06 interpreted the definition of
"no detectable emissions" to require an instrument reading of
zero . The commenter stated that this is incorrect, as other
portions of subpart HHH require an instrument reading of 500
parts per million (ppm) or less for "no detectable emissions."
The commenter provided the following language for revising the
definition of no detectable emissions in subpart HHH:
No detectable emissions means no escape of
Aid ^i wTQ'*-' uio a -^ j. ^^1 LTcTTTTrS
( 1 ) Testiriy Llie device — oi1 system Instrumental
monitoring results in accordance with the requirements
of §63.1282 (d) ; and
(2) No visible openings or defects in the device
or system such as rips, tears, or gaps.
The commenter stated that this comment may also apply to subpart
HH.
2-75
-------
Commenter IV-D-22 recommended the following alternative
definition for no detectable emissions in subpart HH to conform
with the NSPS (subpart KKK) :
Wo detectable emissions means no escape of HAP
from a device or system to the atmosphere as determined
Dy r ( 1 ) x^stJ.1^^ cii£ d4vic£:£> ox £y&t£iki j.n acc!OJ.ci&no£
wlLh .363. 772 (e) the arithmetic difference between the
maximum organic concentration indicated by the
instrument and the background level that is less than
10.000 parts per million by volume; and (2) No visible
openings or defects in the device or system such as
rips, tears, or gaps.
Response; The definition of no detectable emissions does
not require an instrument reading of zero. Therefore, to clarify
the definition of no detectable emissions in final subparts HH
and HHH, the EPA has revised the definition as follows:
No detectable emissions means no escape of HAP
from a device or system to the atmosphere as determined
by:
(1) Testing the device or gyetemlnst rumen t
monitoring results in accordance with the requirements
of §63. 1282 (b) [§63. 772 (c) for subpart HH] ; and
(2) The absence of visible openings or defects in
the device or system such as rips, tears, or gaps.
The EPA believes that restating the requirements of the Test
Methods and Procedures sections in subparts HH and HHH
[§63.771(c) and §63. 1282 (d) ] within the definition of no
detectable emissions is redundant. The EPA did not include these
requirements in the definition.
Process Heater
Comment : Commenter IV-D-22 recommended the following
modification to the definition of process heater in subpart HH to
conform with the ICCR:
Process heater means a device that transfers heat
xi.i3drfl.ccQ. £^y jjULinn tucj. diir€ctJ. to i/i
wj. L» w ^TTwg. L. L. A. ciiiSj L. ^JL^JL .1. vjrm^is o^ttCj^~tHj,&ri WCC^CT
enclosed device using a controlled flame, the primary
purpose of which is to transfer heat to a process fluid
or process material that is not a fluid, or to a heat
2-76
-------
transfer material for use in a process unit (rather
than for steam generation).
Response; Since the definition proposed by the commenter is
reasonable, the EPA has modified the definition of process heater
in subparts HH and HHH, as follows:
Process heater means an enclosed device using a
controlled flame, the primary purpose of which is to
transfer heat to a process fluid or process material
that is not a fluid, or to a heat transfer material for
use in a process (rather than for steam generation)the*
transfers hoat liberated by burning fuel directly to
process streams or to heat tranofer liquids other than
water.
Process Unit
Comment: Commenter IV-G-12 noted that the definition of
natural gas processing plant in subpart HH is identical to that
contained in 40 CFR part 60, subpart KKK, but the definition of
process unit is not included. The commenter was concerned that
without this accompanying definition, subpart HH could be
interpreted ambiguously with respect to exactly what equipment
is, or is not subject to regulatory requirements. The commenter
suggested adding the definition of process unit from subpart KKK.
Response; The term process unit, as used in 40 CFR part 60,
subpart KKK, is necessary to determine the affected sources for
that rule. Since the affected sources in subpart HH are listed
in §63.760(b), the definition of process unit is not necessary.
Production Field Facilities
Comment: Commenter IV-G-01 requested that the EPA clarify
whether outer continental shelf (OSC) platforms are considered
production field facilities.
Response; The determination about whether OCS platforms are
production field facilities is based on existing OCS regulations
and the EPA has not added clarifying language to subpart HH.
Under the current definition, OCS platforms would be considered
2-77
-------
production field facilities when in waters under EPA control as
designated by existing DCS regulations.
Startup and Shutdown
Comment: Commenter IV-D-06 suggested that the EPA have
specific definitions of startup and shutdown in subpart HHH. The
commenter remarked that the EPA's apparent use of the General
Provisions definition of startup and shutdown will not work. The
commenter stated that the definitions in the General Provisions
deal only with startups and shutdowns of a process and do not
deal with startups and shutdowns of control devices or monitoring
systems. The commenter was concerned that the startup, shutdown,
and malfunction plan for a facility would not be applicable for
reducing emissions during a control device or monitoring system
malfunctions. The commenter recommended that the EPA include
specific definitions of startup and shutdown in subpart HHH, and
that those definitions should include control devices and
monitoring systems. The commenter stated that this comment also
applies to subpart HH.
Response; The purpose of the startup, shutdown, and
malfunction plan, as defined in §63.6(e)(3) of the general
provisions (subpart A), is to "ensure that, at all times, owners
or operators operate and maintain affected sources, including
associated air pollution control devices, in a manner consistent
with good air pollution control practices for minimizing
emissions ..." The EPA has added definitions for startup and
shutdown, based on the definitions found in §63.101 of subpart F,
to §§63.761 and 63.1271. It should be noted that the definitions
of startup and shutdown contained in subpart F do not contain
language referring to control devices or monitoring equipment.
State Waters
2-78
-------
Comment: Commenter IV-D-22 recommended adding the following
definition of state waters in subpart HH to clarify the scope of
coverage of offshore facilities:
State waters means those waters for which States
have been granted jurisdiction over offshore lands to a
distance of three nautical miles from their coasts by
the Submerged Lands Act [43 U.S.C. 1301, et seq.]. In
the case of waters offshore from Texas and from Florida
in the Gulf of Mexico, State waters are those waters
over offshore lands for which these two states have
jurisdiction to a distance of three marine leagues
[approximately 10.35 statute miles].
Response; The scope of coverage of OCS platforms is based
on existing OCS regulations. Therefore, the EPA did not add
clarifying language to subpart HH.
Storage vessel
Comment: Commenter IV-D-05 recommended that the EPA define
storage vessel. The commenter asked if a group of tanks at a
facility or each separate tank was considered a storage vessel.
The commenter explained that in the case where two tanks process
900 BPD and share a common vent, applicability could be easily
avoided with by adding vents to each tank and dividing the flow
in half to be below the 500 BPD cutoff.
Response; According to §63.761, a storage vessel is defined
as "a tank or other vessel." Therefore, a group of tanks would
not be considered a storage vessel. Additionally, storage vessel
applicability is on a per vessel basis. In the scenario
described by the commenter, applicability would be based on
actual tank throughput despite vent configuration. Therefore,
the EPA has not made the suggested change.
Storage vessel with the potential for flash emissions
Comment: Commenter IV-D-08 recommended the following new
definition for flash gas in subpart HH:
Flash Gas: VOC emissions from depressurization of
crude oil or condensate when it is transferred from a
2-79
-------
higher pressure to a lower pressure tank, reservoir, or
other container.
Commenter IV-D-07 recommended that the EPA clarify the definition
of flash emissions. The commenter stated that at some
facilities, a small pressure drop exists between a separator and
a storage vessel, so flash emissions are low. The commenter
explained that many facilities dump separators straight into a
wet header system, and have no flash emissions, while other
facilities may only dump low-pressure separators to atmospheric
storage tanks.
Response; The EPA believes that regulating storage vessels
based on tank contents rather than operation will prevent the
possibility of HAP emissions being emitted to the atmosphere via
flashing from uncontrolled tanks. The EPA has added the
following language to the definition of storage vessel with the
potential for flash emissions to clarify what is meant by flash
emissions:
. . .Flash emissions occur when dissolved hydrocarbons
in the fluid evolve from solution when the fluid
pressure is reduced.
Comment: Commenter IV-D-07 suggested that the EPA delete
the requirement for an API gravity of 40° and use the requirement
for hydrocarbon liquids with a GOR of 1,750 scf/bbl of condensate
in the definition of "flash emissions." The commenter stated
that flexibility is added to the definition since if the GOR
changes over time, it could be averaged over one year. The
commenter also suggested that the specific gravity for an API
gravity of 40° is 0.83 which, according to the commenter, seems
"excessively low."
Commenter IV-G-02 stated that controlling storage vessels
with the potential for flash emissions is an appropriate goal
(since most storage tank emissions in oil and gas production are
associated with flash emissions). The commenter explained
however, that using API gravity as a threshold for determining if
2-80
-------
flash emissions occur is not appropriate, by itself, as a good
indication of flash potential. The commenter suggested that GOR
directly measures flash potential and is much more appropriate
for use as a control cutoff criterion. The commenter recommended
that the EPA delete the API gravity criteria from the proposed
definition of "storage vessel with the potential for flash
emissions" in §63.761.
Response; The cutoffs included in the definition of storage
vessels with the potential for flash emissions are intended to
identify storage vessels that have the potential for flash
emissions. The EPA believes that both the API gravity and stock
tank GOR of a liquid are necessary to identify the hydrocarbon
liquids that the EPA believes to have the potential for flash
emissions. In addition, a throughput cutoff of 500 BPD per tank
was added to the definition because the EPA believes that flash
emissions are more likely with higher throughputs. Sections
63.760(a) (1) (iii) and 63.1270(a) (4) of the final rules state that
other parameters used to calculate emissions (such as API gravity
or GOR) must be the maximum for the period over which the maximum
natural gas or hydrocarbon liquid throughput is determined, and
must be based on highest measured values or annual averages. The
EPA has not altered the definition of storage vessel with the
potential for flash emissions based on this comment.
Comment: Commenter IV-D-07 requested that the definition of
storage vessels with the potential for flash emissions clarify
how emission estimates and seasonal operation should be handled.
Response; Since the hydrocarbon liquid throughput cutoff is
based on actual throughput, seasonal fluctuations should not
affect applicability. The EPA has not supplied guidance on
emission estimations in the promulgated rule. The EPA intends to
publish guidance for emission estimations.
2-81
-------
Comment: Commenter IV-D-08 recommended the following
revisions to the definition for storage vessel with the potential
for flash emissions in subpart HH:
Storage vessel with the potential for flash
emissions means any storage vessel that contains a
hydrocarbon liquid with a wellhead-weighted average GOR
equal to or greater than 50 cubic meters (1,750 cubic
feet) per barrel or—and an API gravity equal to or
greater than 40 degrees.
Commenter IV-D-22 recommended the following modification to
the definition of storage vessel with the potential for flash
emissions in subpart HH, to clarify that the GOR is the annual
average wellhead GOR and to indicate the point of measurement:
Storage vessel with the potential for flash
emissions means any storage vessel that contains
hydrocarbon liquids with a GOR equal to or greater than
50 cubic meters (1,750 cubic feet) per barrel
determined as an annual weighted average of the wells
feeding the storage vessel or an API gravity equal to
or greater than 40 degrees measured at the storage
tank.
Response; The GOR is a measure of the amount of entrained
gas contained in a hydrocarbon liquid. Therefore, the higher the
GOR, the higher the potential for flash emissions. The EPA
believes that the GOR should be measured as close to the storage
vessel as possible, as the stock tank GOR, to obtain a realistic
value to determine flash emissions. The EPA has added the phrase
"stock tank" before GOR to the definition of storage vessel with
the potential for flash emissions.
In addition, the EPA has added the throughput cutoff
criterion to the storage vessels with the potential for flash
emissions definition. The final rule states that a storage
vessel with the potential for flash emissions is defined as a
storage vessel that contains a hydrocarbon liquid with a stock
tank GOR equal to or greater than 1,750 scf/bbl and an API
2-82
-------
gravity equal to or greater than 40 degrees, and a hydrocarbon
liquid throughput equal to or greater than 500 bpd. By adding
the throughput criterion to the definition of storage vessels
with the potential for flash emissions, rather than as a cutoff
specified in proposed §63.764(c)(2), storage vessels that do not
meet the criteria for a storage vessel with the potential for
flash emissions are not considered affected sources in the final
rule and are not included in a facility's PTE calculation for
determining major source status.
Surface Site
Comment: Commenter IV-D-16 requested that the word
"platform" be removed from the definition of surface site in
subpart HH. According to the commenter, when covered with the
definition of facility, the definition of surface site could be
misinterpreted to include offshore platforms. The commenter
stated that offshore platforms should not be covered by
section 112 since control of emissions offshore will not protect
the public as there is no public offshore.
Response; It is not the EPA's intent to exempt offshore
platforms. Therefore, the EPA has not removed the term platform
from the definition of surface site.
Comment: Commenter IV-D-22 recommended the following
modification to the definition of surface site in subpart HH to
clarify that individual surface sites connected by linear
installations (e.g., roads, waterways, etc.) are not part of the
same facility:
Surface site means the graded pad, gravel pad,
foundation, platform, or immediate physical location
upon which equipment is physically affixed. Individual
surface sites connected solely by a road, waterway.
walkway, power line, or pipeline shall not be
considered part of the same facility.
Response; The EPA agrees with the commenter and has
modified the definition of facility by adding language to clarify
2-83
-------
that two or more surface sites connected by linear installations
are not part of the same facility. The EPA does not believe it
is necessary to add the same clarifying language to the
definition of surface site. However, the surface site definition
has been revised as follows to specify that graded pad sites and
gravel pad sites are considered surface sites:
Surface site means any combination of one or more
the— graded pad sites, gravel pad sites, foundations.,
platforms., or the immediate physical location upon
which equipment is physically affixed.
Temperature Monitoring Device
Comment : Commenters IV-D-08, IV-D-22, IV-G-02, and IV-G-12
recommended the following modification to the definition of
temperature monitoring device in subpart HH to allow equipment
that uses the Fahrenheit scale, and to remove the accuracy
specifications:
Te/nperature monitoring device means a unit of
equipment used to measure monitor temperature at any
oint in a rocess in degrees F or C arid havin
mwn x t~oi. £i^ &s*.i^i L" c s 5 c d j_n (J / or y \J / wr^xoii^v^L .LS
greater .
Response : The EPA has modified the definition of
temperature monitoring- device, as follows, to allow equipment
that uses the Fahrenheit scale and to clarify that the accuracy
requirements are the minimum allowed to ensure compliance. The
EPA believes that the accuracy requirements are necessary to
ensure that monitoring equipment is operating to demonstrate
ongoing compliance. However, the EPA has modified the level of
accuracy to allow the owner or operator more flexibility in
choosing a monitoring device.
Temperature monitoring device means an instrument
a unit of equipment -used to monitor temperature and
having an minimum accuracy of +2-^ percent of the
temperature being monitored expressed in °C, or +2.5-9-J-5
2-84
-------
°C, whichever is greater
degrees Celsius, or both.
Underground Natural Gas Storage Facilities
Comment: Commenter IV-D-07 stated that the definition for
underground natural gas storage facilities is inconsistent with
the storage process. The commenter stated that the preamble
describes underground storage facilities as "typically extending
from the natural gas processing plant to the local distribution
company." According to the commenter, most storage facilities
recover gas from former production wells, separate the liquids in
high pressure separators, transfer liquids to a wet header system
that transports liquids to a gas processing plant (these liquids
are laden with natural gas), and dehydrates the natural gas
before transfer to the pipeline network.
Response; The preamble to the proposal describes natural
gas transmission and storage facilities as "typically extending
from the natural gas processing plant to the local distribution
company," and underground storage facilities as "subsurface
facilities that store natural gas that has been transferred from
its original location for the primary purpose of load balancing."
It is not the EPA's intent for the description to be all
inclusive of processes at a storage facility and believes that
the description of underground storage sufficiently covers the
storage process. Therefore the EPA has not made any changes to
the regulation in response to this comment.
2-85
-------
2.3 ASSOCIATED EQUIPMENT
Several commenters responded to the EPA's request for
comments on their interpretation of "associated equipment" in
section 112(n)(4) of the Act.
Comment; According to commenters IV-D-02, IV-D-04, IV-D-07,
IV-D-08, IV-D-19, IV-D-20, IV-D-22, IV-D-31, IV-D-33, IV-D-38,
IV-G-02, and IV-G-03,. section 112 (n) (4) mandates no aggregation
of emissions from individual sources at oil and gas production
fields. Commenters IV-D-04 and IV-D-31 stated that the CAA
provides a clear intent to not aggregate emissions from small
sources (e.g., exploration or production wells and their
associated equipment, compressor stations and other similar
units) in order to create major sources. Commenters IV-D-02,
IV-D-07, IV-D-19, IV-D-20, IV-D-33, and IV-D-38 stated that the
EPA exceeded its statutory authority under section 112(n)(4) in
its proposed definition of facility, by allowing for the
aggregation of emissions from glycol dehydration units and
storage vessels with the potential for flash emissions. The
commenters requested that the EPA be consistent with the CAA.
According to commenter IV-D-20, the EPA did not create a
regulatory definition that is true to the statute.
Response; The EPA disagrees that the Agency exceeded its
statutory authority for the reasons discussed below. Section
112(a)(1) generally requires HAP emission points within a
contiguous area and under common control to be aggregated in a
major source determination for the purposes of section 112.
While this approach is appropriate for facilities in most
industries, it may lead to unreasonable aggregations if strictly
applied to oil and natural gas field operations. Given that some
oil and natural gas operations (e.g., a production field) may
cover several square miles or that leases and mineral rights
agreements give some companies control over a large area of
contiguous property, determination of major source status
2-86
-------
strictly by the language of section 112(a)(1) could mean in this
industry that HAP emissions must be aggregated from emissions
points separated over large distances.
Congress addressed the unique aspects of the oil and natural
gas production industry by providing the special provisions in
section 112(n)(4) of the Act referring to the ". . .oil and gas
exploration or production well (with its associated
equipment)..." However, Congress did not provide a definition
of the term "associated equipment" in the statutory language,
leaving its interpretation to the EPA. A definition of this term
is important in determining the major source status of facilities
in both the oil and natural gas production and the natural gas
transmission and storage source categories.
In the absence of clear guidance in the statute, the EPA
evaluated various options for defining "associated equipment"
prior to proposal and developed the proposed definition. The
conunenters did not offer substantive new information to support
their claim that the EPA had exceeded its authority. The next
comment and response provide additional information regarding the
development of the definition of associated equipment.
Comment: Commenters IV-D-07, IV-D-19, IV-D-20, IV-D-33, and
IV-D-38 did not agree with the EPA's definition of associated
equipment, and contended that glycol dehydration units and
storage vessels with the potential for flash emissions should be
considered "associated equipment."
Although they did not support the EPA's proposed definition
of associated equipment, commenter IV-D-20 stated their support
for the EPA's efforts to control HAP that represent an adverse
impact to human health and the environment by focusing on the
sources that emit the most HAP from the oil and natural gas
production source category.
2-87
-------
Commenters IV-D-19 and IV-D-20 were concerned that creating
exemptions from the terms of the statute would create a negative
precedent for this and future rules and, along with commenter
IV-D-38, requested that the EPA modify the definition of major
source and associated equipment to comply with the provisions of
CAA section 112(n)(4)(A).
Commenter IV-D-29 supported a more narrow definition of
"associated equipment" that also excludes glycol dehydration
units and storage vessels with flash emissions. The commenter
was concerned that many potential major sources in Santa Barbara
and Ventura counties would be exempted under the broad definition
of associated equipment.
Although commenters IV-D-08, IV-D-22, IV-G-02, and IV-G-03
did not fully support the EPA's interpretation of "associated
equipment," they acknowledged that the proposal to limit the
aggregation of emissions to HAP from glycol dehydration units and
storage vessels with the potential for flash emissions was a
workable solution to the aggregation of all HAP sources. The
commenters suggested that the aggregation of dehydration units
and storage vessels with flash potential would result in the same
major source determination as aggregation of all potential
sources, but would reduce the burden on the facility operator.
In addition, commenter IV-D-05 stated that the rationale and
conclusions used to clarify ancillary equipment for aggregating
emissions for a major source determination seemed appropriate.
[Note: The commenter mentioned "ancillary equipment" but most
likely meant "associated equipment," as the comment appears to be
directed toward associated equipment.] According to the
commenter, the EPA's decision to aggregate glycol units and
storage vessels with flash emissions will "give the rule some
degree of effectiveness."
Response; According to the statutory definition of major
source in section 112(a)(1) of the Act, HAP emissions from all
emission points within a contiguous area and under common control
must be counted in a major source determination. By stating that
2-88
-------
emissions from any oil and gas production and exploration well
(with its associated equipment) cannot be aggregated for a major
source determination, the provisions of section 112(n)(4)(A) mean
HAP emissions from each well and each piece of equipment
considered to be associated with the well must be evaluated
separately in a major source determination. That is, any well or
piece of associated equipment would only be determined to be a
major source if HAP emissions from that well or piece of
associated equipment were major.
Therefore, to implement this special provision of the Act
for the oil and natural gas production source category, a
definition of "associated equipment" was necessary. The EPA
proposed that "associated equipment" be defined as all equipment
associated with a production well up to the point of custody
transfer, except that glycol dehydration units and storage
vessels with the potential for flash emissions would not be
associated equipment. In developing this proposed definition,
the Agency evaluated several options. The Agency also sought and
received input from industry and other stakeholders.
In the absence of clear guidance in the statute, the EPA
evaluated various options for defining "associated equipment"
prior to proposal. The EPA's objective was to arrive at a
reasonable interpretation that would: (1) provide substantive
meaning to the term "associated equipment" consistent with
congressional intent; (2) prevent the aggregation of small,
scattered HAP emission points in major source determinations;
(3) be easily implementable; and (4) not preclude the aggregation
of significant HAP emission points in the source category. Due
to the lack of clarity in the statute and the potential impact on
major source determinations, the Agency worked with industry
2-89
-------
stakeholders to identify and evaluate options prior to proposal.
Industry representatives expressed their goals for the
interpretation of associated equipment, and provided information
on the magnitude of HAP emissions points and the potential
impacts of various options considered by the EPA.
The EPA considered, but rejected a definition based on a
narrow interpretation that would include only valves and fittings
on a well as being associated equipment primarily because this
option would not provide any additional relief to industry beyond
what would have been provided had Congress only used the term
"well" in section 112(n)(4) of the Act. The EPA also rejected a
definition, initially recommended by industry, that was based on
a broad interpretation that would include equipment far beyond
the well as associated equipment.
In discussions with industry stakeholders over an extended
period of time prior to proposal, the Agency sought to reach a
workable solution on the definition of associated equipment, one
that recognized the need to implement relief for this industry as
Congress intended, and that also allowed for the appropriate
regulation of significant emission points. In a technical
evaluation, the EPA identified glycol dehydration units and
storage tanks with flash emission potential as substantial
contributors to HAP emissions, particularly relative to sources
such as production wells. This conclusion was supported by
industry. Under the proposed approach, associated equipment was
defined as all equipment up to the point of custody transfer,
excluding glycol dehydration units and storage vessels with the
potential for flash emissions. This approach also included a
definition of facility in the rule that effectively limited the
distance over which all emission points (including glycol
2-90
-------
dehydration units and storage vessels with the potential for
flash emissions) may be aggregated. Based on discussions with
industry prior to proposal, as well as comments received
supporting the proposed definition of associated equipment, the
Agency believes that the proposed approach best meets both
industry and EPA goals for implementation of the language of
section 112(n) (4) .
Commenters who argued that the Agency exceeded its authority
with the definition of associated equipment offered no
substantive new information to support their claim. The EPA
could not find support in the statute or in the legislative
history10 that indicated that Congress intended to preclude
aggregation of all emission points, including such significant
ones as glycol dehydration units and storage tanks with flash
emission potential through their inclusion as associated
equipment. Rather, there are clear indications, in the EPA's
judgement, that Congress1 primary intent was to preclude the
aggregation of small emitting sources over vast distances. The
legislative history of the Act, for example, indicates that
Congress believed that oil and natural gas production wells and
their "associated equipment" generally have low HAP emissions,
and are typically located in widely dispersed geographic areas,
rather than being concentrated in a single area. The EPA used
this background as a guide in developing an interpretation of
"associated equipment" along with available data on HAP emissions
from emission points within the oil and natural gas production
source category. The EPA believes that glycol dehydration units
10
Conference Debates and Report. In: A Legislative History
of the Clean Air Act Amendments of 1990. U.S. Government
Printing Office. November 1993. P. 1238.
2-91
-------
and storage vessels with the potential for flash emissions are
not the type of small HAP emission points that Congress intended
to be included in the definition of associated equipment.
After the EPA's review and consideration of all comments
received on the proposal, the definition of associated equipment
promulgated in today's rule is the same as proposed.
Comment; Commenter IV-D-07 remarked that it is arbitrary to
include storage vessels in which treating and processing does not
take place as associated equipment, because these vessels have
the highest potential for flash emissions as compared to storage
vessels further downstream.
Response; With regard to including, as associated
equipment, storage vessels in which no treating or processing
takes place, this statement in the preamble to the proposal
referred to an intermediate option that the EPA considered.
Under this option, all equipment up to the point where initial
processing of an extracted hydrocarbon stream takes place would
be considered associated equipment. Thus, only the Christmas
tree and storage vessels in which no treating or separation takes
place would be considered associated equipment. However, this
option was rejected by the EPA as being a definition that was too
narrow.
The option selected by the EPA states that storage vessels
with the potential for flash emissions are not to be considered
associated equipment. Therefore, the EPA has focussed the
standards (subpart HH) on those storage vessels with significant
emissions.
Comment: Commenter IV-G-01 requested that the EPA clarify
whether "all equipment from the wellbore to the point of custody
transfer" (as stated in §63.761) includes "ancillary equipment."
2-92
-------
Response; Associated equipment is defined in §63.761 to
include all equipment from the wellbore to the point of custody
transfer, except glycol dehydration units and storage vessels
with the potential for flash emissions. Therefore, prior to the
point of custody transfer, ancillary equipment is considered to
be associated equipment and cannot be aggregated to determine
major source status. Therefore, the EPA has not made any changes
to subpart HH in response to this comment.
Comment: Commenter IV-D-07 felt that inserting the
statement about custody transfer does not clarify the meaning of
associated equipment. The commenter explained that custody
transfer usually occurs after a preliminary separation of gas and
liquids, which includes the use of storage vessels with flash
emissions.
In contrast, commenter IV-D-05 stated that the term after
custody transfer is "probably the most universal term that can be
used in regards to clarity."
Response; As stated in a previous response, the EPA defined
associated equipment to comply with its interpretation of section
112(n)(4)(A) of the CAA. Although storage vessels with the
potential for flash emissions typically occur prior to the point
of custody transfer, the EPA specifically excluded these
emissions sources from the definition of associated equipment in
an effort to focus on significant HAP emission points.
The term "custody transfer" is included in the definition of
associated equipment as the point of delineation for where
emission aggregation of all emission points within a facility may
occur. It is also used as the basic point to define where the
oil and natural gas production source category ends and the
natural gas transmission and storage source category begins. The
exceptions to this are natural gas processing plants, which are
2-93
-------
included in the scope of the oil and natural gas production
source category even though they are considered to be after a
point of custody transfer. This exception was provided because
natural gas processing plants are typically clearly defined
facilities within this source category. Furthermore, during the
development of the proposed regulations, industry repeatedly
stated that the term custody transfer is well understood within
the industry and that custody transfer of hydrocarbon streams
occurs only once and not multiple times. Therefore, the EPA has
not made any changes to the definition of associated equipment in
response to this comment.
2-94
-------
2.4 HAP EMISSION POINTS
Comment: The EPA specifically requested information and
comments, along with supporting documentation, on HAP emissions
from several emission points. These emission points included:
(1) process vents at amine treating units and sulfur plants,
(2) transfer and storage of pipeline pigging wastes,
(3) combustion sources located at oil and natural gas production
and natural gas storage and transmission facilities, and
(4) storage vessels at natural gas transmission and storage
facilities.
Amine Treaters and Sulfur Units. Commenters IV-D-07,
IV-D-16, IV-D-22, and IV-G-09 responded to the EPA's request for
information on amine treaters and sulfur units. The commenters
stated that amine treaters are most likely small sources of HAP
and there is little available data to estimate HAP emissions from
these sources. Commenter IV-D-07 requested clarification on how
these emissions should be estimated. Commenter IV-D-16
recommended that if amine units and sulfur recovery units are
shown to be significant sources of HAP, they can be controlled
during the residual risk review required by section 112 (f) of the
CAA. Commenter IV-D-22 stated that since there are relatively
few units (as compared to glycol dehydration units and storage
tanks), the amine treater unit and sulfur recovery unit totals
would result in an extremely small percentage of total baseline
year HAP for the source category. Commenter IV-G-09 stated that
amine plants are designed to remove carbon dioxide, sulfur, and
other impurities from the gas. According to the Commenter, most
aromatic and long chain hydrocarbons are removed from natural gas
for their economic value before the gas treatment in amine
plants. The commenter further explained that the large amount of
non-combustible gases in the vents of amine plants makes flaring
impractical and the high flow rate of these non-condensable gases
makes condensers technically not feasible. At this time, the
commenter stated that they were not able to provide an example of
a practical control technology and recommended not regulating
amine plants under this standard.
2-95
-------
Pipeline Pigging Operations. Commenters IV-D-07, IV-D-08,
IV-D-10, IV-D-16, IV-D-17, IV-D-20, IV-D-22, and IV-G-09
responded to the EPA's request for information on pipeline
pigging operations. According to the commenters, pipeline
pigging activities are performed to remove scale and other
accumulations, and occur intermittently and infrequently with
insignificant fugitive emissions. 'Commenter IV-G-09 stated that
although there is no set schedule, most transmission lines are
pigged less than once per year, and are open to the atmosphere
for only the few hours required to discharge the liquids and the
solids collected. The commenters stated that pigging wastes are
contained in storage tanks and have minimal emissions. The
commenters requested that the EPA not regulate these sources.
According to commenters IV-D-07, IV-D-08, IV-D-20, IV-D-22, and
IV-G-09 the wastes generated from pigging are primarily solids
with entrained liquids and contain small amounts of VOC and HAP.
The commenters concluded that potential HAP emissions from
pipeline pigging operations would be inconsequential in the oil
and natural gas production source category.
Commenter IV-D-10 suggested that, until the EPA has
developed specific requirements and applicability determinations
for HAP emissions from transfer and storage of pipeline pigging
wastes and combustion sources, these units should not be covered
under the oil and gas MACT. The commenter was mostly concerned
with estimating PTE for HAP emissions from transfer and storage
of pipeline pigging wastes and combustion sources from the
gathering portion of their company. Besides the infrequent
occurrence of pigging operations, the commenter stated that there
is no testing method available for measuring these emissions from
the pigging waste storage tanks. According to the commenter,
quantifying emissions from the tanks is difficult. In addition,
the flow is difficult to measure given the unsteady flow
conditions and the variability of the gas composition over time.
The commenter stated that it would be unfair to the industry to
assume they are major sources due to emissions from pigging since
2-96
-------
there is no ability to test for applicability to the
requirements.
Combustion sources. Commenters IV-D-20 and IV-D-22
discussed combustion sources. The commenters suggested they were
not aware of any existing database that adequately characterizes
the populations of equipment, HAP emissions, or risk of exposure
to the public for combustion sources at oil and gas production
facilities. The commenters also noted that the ICCR had been
initiated, by the EPA, to address combustion sources. Commenter
IV-D-20 also reminded the EPA that combustion sources in the
Appalachian region are unique from those in the Southwest. The
commenter explained that in Appalachia, the oil and gas
production related combustion sources are generally in non-urban
areas, emit small amounts of HAP and pose little if any risk to
human health.
Storage vessels at natural gas transmission and storage
facilities. Several commenters responded to the EPA's request
for information on storage vessels from natural gas transmission
and storage facilities. Commenters IV-D-08, IV-D-12, IV-D-16,
IV-D-35, and IV-G-12 suggested that there are small amounts of
liquids associated with transmission and storage facilities. The
commenters also stated that the liquids associated with
transmission and storage facilities contain small amounts of HAP,
resulting in insignificant emissions. Commenter IV-D-07 stated
that at some facilities, a small pressure drop exists between a
separator and a storage vessel, so flash emissions are low.
According to the commenter, many facilities dump separators
straight into a wet header system, and have no flash emissions,
while other facilities may only dump low-pressure separators to
atmospheric storage tanks. The commenter and commenter IV-D-31
recommended that an annual average be used to evaluate "flash
potential" if the EPA decides to regulate storage vessels in the
natural gas transmission and storage category. Commenter IV-D-08
explained that by the time natural gas enters the transmission
facility, most of the liquid has been removed. The commenter
further explained that the entrained liquids are collected in
2-97
-------
barrels and emptied infrequently. According to the commenter,
requiring controls on storage vessels at natural gas transmission
and storage facilities would provide negligible HAP emission
reduction. Commenter IV-D-12 recommended that these sources not
be subject to emission controls unless, and until, the EPA has
collected and analyzed adequate information to demonstrate that
controls are justified. Commenter IV-D-16 suggested that if the
applicability sections of subpart HH and HHH do not overlap, the
tanks that need control will be covered under subpart HH.
Commenter IV-D-35 stated that very few natural gas transmission
and storage facilities in the State of Colorado use storage
vessels. Commenter IV-G-12 remarked that storage tanks at their
gas storage facilities are used to hold lubricating and fuel oils
for internal combustion engines, or fluids with very low vapor
pressures such as glycol or produced water. According to the
commenter, the calculation of emissions from these tanks for
title V permitting purposes showed that they are of a de minimis
nature.
Response; Based on the comments received, the EPA believes
that process vents at amine treating units and sulfur plants,
transfer and storage of pipeline pigging wastes, combustion
sources located at oil and natural gas production and natural gas
storage and transmission facilities, and storage vessels at
natural gas transmission and storage facilities are not
significant sources of HAP and do not warrant regulation under
subparts HH and HHH. If warranted, combustion sources at natural
gas transmission and storage facilities may be regulated under
future regulations.
However, a recently published report from GRI addressed HAP
emissions from amine treater and sulfur recovery units. The
report indicated that amine treaters and sulfur units are not
significant contributors to overall national HAP emissions from
2-98
-------
the oil and natural gas production source category. However, the
report indicated that amine treaters may be significant
contributors to HAP emissions on a site-specific facility basis.
It should be noted that emissions from amine treaters and sulfur
recovery units must be taken into consideration by a facility's
owner or operator in making major source determinations.
2-99
-------
2.5 IMPACTS
2.5.1 Cost Impacts Including Production Recovery Credits
Comment: Commenters IV-D-07 and IV-D-12 stated that the
cost data are not representative of the true impact of subpart
HHH and demonstrates that the data base is inadequate. The
commenters referred to the economic analysis which indicated that
only five facilities would be affected at a collective capital
cost of $57,000. According to the commenters, the size range
examined was 20 to 50 MMscf/d, which is not at all representative
of the dehydration equipment they operate. Commenter IV-D-12
reviewed the proposal and determined that at least four major
sources on their system would be subject to subpart HHH.
Additionally, the commenter provided an example of condenser
controls equivalent to those proposed in the standard that had
been installed on two dehydration units of 300 MMscf/d each, at a
single site, at a capital cost greater than $500,000. The
commenter remarked that this example demonstrates that the EPA's
understanding of the natural gas transmission and storage source
category is flawed and deficient. The commenter recommended that
the EPA take more time to understand the industry, and to analyze
emissions and cost impacts before proceeding further with the
rulemaking. Commenter IV-D-07 also mentioned that the
recordkeeping costs seem very low.
Commenters IV-D-15 and IV-G-05 stated that based on their
experience, the EPA has underestimated the cost of installation
of condensers and monitoring equipment. As an example, commenter
IV-D-15 stated that a recently purchased natural draft condenser
on a 5.0 MMscf/d glycol dehydrator unit cost more than $14,000,
not including the cost of installation, tanks, and temperature
monitoring equipment. The temperature monitoring equipment (two
temperature sensors and a chart recorder) cost $2,000. The cost
for an installation using an existing tank was closer to $18,000
and did not include costs for additional controls (e.g., a
flare). The commenter, along with commenter IV-G-05, provided
another example of a 20.0 MMscf/d unit where the cost of the
condenser and connections to allow combustion of the vent offgas
2-100
-------
in the glycol reboiler was more than $22,000 for the equipment,
without installation costs or temperature monitoring equipment
costs. The commenters compared their numbers to the costs in the
BID of $11,000 (Table 6.1) for a comparable control scheme
including a new tank for a comparable unit. Commenter IV-G-05
maintained that the costs of condensers and monitoring equipment
may exceed the value of the gas being treated, and some units
will probably shut down to avoid the cost of installing and
maintaining this equipment.
Response; The EPA based its cost estimates on published
installed control system costs from the Ventura County
(California) Air Pollution Control District (APCD). These costs
were associated with a glycol dehydration unit regulation issued
by the Ventura County APCD (Air Docket A-94-04 number IV-A-07).
According to this information, the cost of installing a condenser
control system does not vary significantly based on the size
(capacity) of a glycol dehydration system.
However, to address comments received from the natural gas
transmission and storage source category, the EPA collected
additional data from 83 glycol dehydration units located at
natural gas transmission and storage facilities. This additional
information, as well as the information on 31 glycol dehydration
units collected for the proposal, indicated that 71 glycol
dehydration units were controlled. The EPA determined that 59 of
the 71 glycol dehydration units were controlled using a
combustion device, primarily a flare. Based on this new data,
the EPA revised the cost impacts for the natural gas transmission
and storage (Air Docket A-94-04 number IV-A-08). The EPA
estimated that seven facilities would be affected by subpart HHH.
The EPA assumed that six of the facilities would install flares
to meet the control requirements, and one facility would install
2-101
-------
a condenser. Therefore, based on these control scenarios,, the
EPA estimated a total capital investment of $280,000 and a total
annual cost of $300,000 per year for the natural gas transmission
and storage source category. This annual cost estimate includes:
(1) the cost of capital, (2) operating and maintenance costs,
(3) the cost of monitoring, recordkeeping, and reporting fMRR) ,
and (4) any associated product recovery credits.
Comment; Commenter IV-D-15 stated that, in their
experience, the cost for implementing a leak detection and repair
(LDAR) program for a natural gas processing plant would cost
approximately $6.50 to $7.50 per component monitored for the
first year and $5.00 to $6.00 per component monitored for
subsequent years. According to the commenter, remote locations
will add to such costs, and these costs do not include repair
work. The commenter calculated first year costs ranging from
$2,600 for the 400 components monitored for the Model "A" Plant
up to $17,250 for the 2,300 components for the Model "C" Plant.
The commenter compared their costs to the $400 that is provided
in the example in the BID.
Response; The EPA estimated that the total annual costs for
LDAR programs range from $12,000 (in July 1993 dollars) for model
natural gas processing plant "A" to $42,000 for model natural gas
processing plant "C." These costs are documented in a memorandum
contained in the docket (Docket Item II-A-03) .ll The $400 (as
shown in Table C-3 of the Background Information Document)
represents additional MRR costs that have not been previously
accounted for in the LDAR program costs.
Comment: Commenter IV-D-10 stated that the EPA did not take
into account the additional costs to companies to dispose of
nG. Viconovic, EC/R Inc., to M. Smith, EPA-.WPCG, and L.
Conner, EPA:ISEG. Cost impact estimates for the oil and natural
gas production and natural gas transmission and storage national
emission standards for hazardous air pollutants. July 9, 1996.
2-102
-------
condensed water as exempt waste. According to the commenter, the
disposal costs in the San Juan Basin of New Mexico are
approximately $1.40 per barrel of water. The commenter stated
that condensing will at least double the water disposal amounts
from dehydrators because steam currently going to the atmosphere
will be condensed.
Response; Approximately 20 billion barrels per year of
produced water are generated by the oil and natural gas
production source category.12 Using GLYCalc to determine the
amount of produced water generated by the number of facilities
estimated to be affected by subpart HH, the EPA calculated that
the oil and natural gas production NESHAP would result in an
increase in produced water production by approximately 590,000
barrels per year. According to a GRI report,13 produced water
would be typically handled along with other produced water
streams, either by underground injection control, surface
impoundment, or other miscellaneous methods. Thus, the EPA
believes that the oil and natural gas production NESHAP would
have a minimal impact on existing produced water disposal costs
and control costs.
Comment: Commenter IV-D-22 stated that the EPA has
substantially understated the cost of subpart HH to industry
sources and has underestimated the monitoring, reporting, and
recordkeeping burdens associated with the rule. The commenter
estimated the capital costs of subpart HH to exceed $25 million
12The Oil and Gas Exploration and Production Industry.
Trends: 1985-2000. (Draft Report, April 30, 1993). U.S. EPA,
Washington, DC. (Air Docket A-94-04 number IV-A-09).
13Rueter, C.O., M.C. Murff, and C.M. Beitler (Radian
International LLC). Glycol Dehydration Operations, Environmental
Regulations, and Waste Stream Survey. Prepared for the Gas
Research Institute. Publication Number GRI-96/0049. June 1996.
Page 4-38.
2-103
-------
for major sources, as compared to the EPA's estimates of $6.5
million for major sources. The commenter also estimated the
annual costs of subpart HH to be $15 million for major sources,
as compared to the EPA's estimates of $4 million for major
sources. The commenter estimated cost effectiveness to be
$3,000/megagram for the EPA's model plant, as compared to the
EPA's cost effectiveness of $116/megagram for its model plant.
The commenter mentioned the EPA's request for comments on
the production recovery credit assumed to result from
installation of control devices. The commenter believes, based
on GLYCalc runs, that the EPA has materially overstated the
quantity of product recovered that could be sold to offset the
capital and annual costs associated with subpart HH.
Response; The EPA based its national cost estimate impacts
on the estimated number of facilities that would be impacted by
the regulatory provisions of subparts HH and HHH, along with
detailed emission control cost estimates per HAP emission point.
In addition, the MRR costs were based on a detailed analysis of
the regulatory requirements of subparts HH and HHH. The EPA
believes that the MRR cost estimates reflect the estimated effort
required to address MRR requirements in the final subparts HH and
HHH.
In reviewing the costs presented by the commenter, the EPA
noted that the commenter compared a 10 MMscf/d unit with a
35 MMscf/d unit, which the EPA used as its example cost model
plant in Chapter 6.0 of the BID. Thus, the commenter observed a
significant difference in product recovered.
Comment: Commenter IV-D-38 made general statements that the
proposed regulations will have a profound impact on their
operations. This includes a cost of implementation that is
enormous compared with environmental benefits and, ultimately,
passing these increased costs to consumers.
2-104
-------
Response; The EPA also conducted an economic impacts
analysis to evaluate the impacts of the regulation of affected
producers, consumers, and society (Air Docket A-94-04 numbers
II-A-08 and IV-A-13). The EPA estimated that price and output
changes as a result of the regulation were less than 0.01 of 1
percent, which is significantly less than observed market trends
(based on 1992 and 1993 data). The cost impacts are presented in
tables 1 through 3 of the preamble; the development of these
costs is documented in the background information document.14
Through the comment period, the EPA provided the opportunity for
the public to submit comments, along with supporting
documentation, on all aspects of the proposed NESHAP. Without
supporting documentation to address the specific impacts of the
proposed NESHAP on the commenter's operations, the EPA is unable
to specifically respond to this comment.
Comment: Commenter IV-G-09 estimated the average cost of
the condenser and auxiliary equipment it placed on dehydration
units, each processing 100 MMscf/d, to be more than $150,000, in
contrast to the EPA's preamble statement that the average cost of
condenser control devices would be less than $12,000.
Response; As previously stated, the EPA based its cost
estimates on published installed control system costs from the
Ventura County APCD (Air Docket A-94-04 number IV-A-07). Without
substantive supporting documentation to address the specific cost
impacts of the NESHAP on the commenter's operations, the EPA is
unable to specifically respond to this comment.
""National Emissions Standards for Hazardous Air Pollutants
for Source Categories: Oil and Natural Gas Production and Natural
Gas Transmission and Storage - Background Information for
Proposed Standards." EPA-453/R-94-079a. April 1997.
2-105
-------
Comment: Commenter IV-D-11 stated that imposing
requirements on sources that are already controlled invalidates
the EPA's cost effectiveness analysis. According to the
commenter, sources controlled under the Louisiana Department of
Environmental Quality (LDEQ) Section 2115 Waste Gas Disposal Rule
already meet the MACT floor efficiency requirements. Therefore,
the commenter stated that the expenditures for monitoring and
recordkeeping requirements will not result in appreciable
emissions reductions. The commenter concluded that the cost
effectiveness for controlled sources could approach infinity
since no additional reductions will be realized. The commenter
referred to the State of Louisiana's implementation of
significant controls on glycol dehydrators and provided a copy of
the pertinent pages of the LDEQ 1995 Annual Report, which shows
that toxic air emissions from glycol dehydrators have been
reduced from 36,720,000 pounds (Ib) to 1,277,608 Ib in 1994,
representing a reduction greater than 96 percent. According to
the commenter, further reductions have occurred since 1994. The
commenter explained that the great majority of the emissions
reductions are federally enforceable since they were required due
to the Louisiana Administrative Code (LAC) 33:111.2115 Waste Gas
Disposal Rule and are contained in either State-only or part 70
permit programs in LAC 33:III.Chapter 5. The commenter noted
that these rules are incorporated in the EPA-approved State
implementation plan (SIP) for Louisiana and that further
reductions are being accomplished due to a rule for glycol
dehydrators (LAC 33:111.2116) and for flash gas from storage
tanks (LAC 33:111.2104, Crude Oil and Condensate rule) which are
not included in the EPA-approved SIP.
Response; The commenter seems to imply that the
installation and operation of control equipment are all that is
necessary to achieve the MACT floor efficiency requirement. The
EPA believes that monitoring, recordkeeping, and reporting
requirements serve a vital function in ensuring that an emission
2-106
-------
limitation is initially, and continues to be, met. Therefore,
the EPA maintains that the costs associated with monitoring and
recordkeeping are part of the costs that must be incurred to
achieve the necessary emission reduction/ not additional costs
with no appreciable emission reduction.
The monitoring, recordkeeping, and reporting requirements
contained in the final rule are the requirements that the EPA
believes are necessary to ensure compliance with the emission
limitations. If the Louisiana rule referred to by the commenter
requires comparable monitoring, recordkeeping, and reporting as
that contained in the final rule, then no additional burden would
be incurred by affected facilities. In fact, via §63.10(a)(3) of
the General Provisions, which is incorporated by reference into
the final rule, an owner or operator can simply send the
Administrator a copy of any report to the State that contains the
same information required by the federal NESHAP. If the
monitoring, recordkeeping, and reporting currently being
conducted by these Louisiana facilities do not meet the
provisions in the final rule, then the EPA believes that these
sources would need to upgrade efforts to ensure compliance with
the MACT standard.
Further, if the State of Louisiana believes that the rule
referenced by the commenter is equivalent to the final standard
for Oil and Natural Gas Production facilities, then an
application could be made under subpart E of 40 CFR 63. If the
EPA agrees that the Louisiana rule is at least as stringent as
the federal rule, the State rule would replace the federal NESHAP
for source in the State of Louisiana.
2.5.2 Environmental Impacts
2-107
-------
Comment: Comtnenter IV-D-07 stated that the EPA's
assumptions made in estimating emissions are not representative
of their operations, especially HAP concentrations in natural gas
before dehydration. The commenter noted that methane and
volatile organic compounds (VOC) emission reductions are included
in the preamble even though the standard is designed to reduce
HAP emissions. The commenter questioned why methane emissions
were included.
Response; The EPA used the best available information to
estimate the environmental impacts of subparts HH and HHH. The
information represents average facilities and operating practices
within the source categories and not any individual facility and
facility operations. The EPA included VOC and methane emission
reductions for the proposed regulations for informational
purposes and to show the overall emission reduction benefits
associated with these NESHAP. Methane and VOC reductions were
not used to set the standards for subparts HH and HHH. Therefore
the EPA has not made any changes to subparts HH and HHH in
response to this comment.
Comment: Commenter IV-D-07 stated that the estimates for
increases for nitrogen oxides and sulfur dioxide emissions from
additional flare operations may be severely underestimated.
Response; The EPA used AP-42 emission factors to estimate
the increases in nitrogen oxides and sulfur dioxide emissions
associated with the installation of flares at certain remote
facilities. The EPA believes these estimates are representative
of potential emission increases for this industry. Therefore,
the EPA has not made any changes to these estimates.
2-108
-------
2.6 ECONOMIC ANALYSIS
Comment: Commenter IV-D-19 was concerned that any analysis
of the economic impact of the proposed standard should adequately
consider the impact on marginal production operations.
Specifically, the commenter mentioned the low crude oil prices
and high level of abandonment of marginal oil production
operations. The commenter indicated that to meet the
requirements of the Regulatory Flexibility Act (RFA), the EPA
screened a sample of small entities and determined that minimal
impacts from subpart HH would result. The commenter requested
assurance or modification of the economic impact analysis to
ensure the screening sample contains an appropriate cross section
of small entities.
Response; The Agency's economic analysis employed a
baseline characterization of the industry that includes marginal
production operations. This baseline characterization linked the
model plants and units developed by the EPA's engineering
analysis with the well groups identified and characterized by the
Gruy Engineering Corporation in their 1991 report prepared for
API. These well groups are defined by production rates in
specified ranges of well depths for both oil and gas wells in
each of the 37 different geographic areas across the United
States. The Energy Information Administration (EIA) report
provides details for oil well groups in Appendix A and for
natural gas well groups in Appendix B. Therefore, to the extent
that the Gruy Engineering Corporation's database appropriately
characterizes marginal production operations, the Agency's
economic analysis includes the impacts on these operations.
The Agency expects that the impacts on these marginal
operations will be minimal given the size cutoff for glycol
dehydration units. Glycol dehydration units that process less
than 3 MMscf/d are not affected by the proposed standards. It is
2-109
-------
likely that a large share of the marginal operations operates
these smaller units and, thus, would not incur compliance costs
associated with the proposed standard. Furthermore, it follows
that the smaller owners would likely own only units of this type
and, thus, would also not be adversely affected by the proposed
standard. However, in accordance with the RFA, the Agency still
conducted an analysis of the small business impacts of the
standards. As noted by the commenter, the Agency employed a
sample of companies to evaluate the small business impacts
because the necessary financial data were not available for each
and every potentially affected company. The sample of 80
companies was determined by data availability and considered a
fair representation based on their distribution across the
relevant SIC codes. To facilitate evaluation of the
appropriateness of the sample, the EIA report provided a list of
the companies that were part of the sample as well as their
baseline data in Appendix F. Based on Small Business
Administration (SBA) size standards, the EPA's sample contained
39 small companies, which were 48.8 percent of all companies.
Based on this sample, the Agency determined that the mean
cost-to-sales ratio for small companies was 0.1 percent with a
maximum of 1.1 percent. This information supports the Agency's
finding that there will not be a significant impact on a
substantial number of small entities.
Comment: Commenter IV-D-24 stated that the EPA economic
analysis appears to significantly underestimate the control costs
of the proposed regulation. The commenter estimated that
post-regulatory control costs could be $121.2 million/yr as
compared to the EPA's estimate of $18.9 million/yr. [Note:
These figures include the costs associated with the regulation of
2-110
-------
area sources, which has been deferred until the development of
the Urban Air Toxics Strategy is finalized.]
Furthermore, the commenter estimated that as many as 1,960
wells per year could be abandoned as a result of the increased
compliance costs of the proposed regulation and the EPA estimates
that no wells would be closed prematurely. The commenter
explained that the EPA's determination of abandonment was based
on aggregate changes in corporate revenues and profits. However,
the commenter stated that production decisions are made on a
well-by-well or project basis and if an individual project's
profits fall below its break-even point, that well will be
abandoned.
The commenter also estimated that an average of 46,000
thousand cubic feet (MCF) of natural gas production could be lost
each year as a direct result of the increased costs of the
proposed exploration and production (E&P) MACT regulation,
compared to the EPA's estimate of 99 thousand cubic feet per year
(MCF/yr). The commenter stated that their sophisticated,
field-specific benefit-cost model and a detailed gas supply
model, to estimate production impacts, provides a more accurate
estimate than the EPA's.
The commenter noted that the EPA did not estimate losses of
economically producible natural gas reserves. The commenter
estimates that reserve losses could average as much as 1,040
billion cubic feet per year (bcf/yr) through 2010. The commenter
stated that while employment impacts of the proposed rules are
estimated to be minimal in their analysis, the EPA should update
its analysis to include this employment loss.
Response; The Agency's engineering analysis, as summarized
in Section 3 of the EIA report, has estimated the annual
compliance cost of the proposed standard to be almost $19 million
per year, with major sources incurring $7 million annually and
area sources incurring $12 million per year. The Agency's
economic impact results are based on this estimate of the annual
2-111
-------
compliance cost of the proposed standard. This estimate differs
significantly from the commenter's estimate of $121.2 million per
year. Therefore, it is not surprising that the commenter's
reported economic impacts are much greater than those reported by
the Agency. The differences in these economic impacts are
attributable to the significant disparity in the cost estimate
used in determining these economic impacts as opposed to the
economic methodology. The Agency expects that input of these
higher compliance costs in its model would likely provide more
comparable impacts to the commenter's "sophisticated" economic
model. In addition, it is not clear whether the coxnmenter has
accounted for the EPA's size cutoff of 3 MMscf/d for TEG
dehydration units in computing its economic impact results. The
Agency expects that this size cutoff prevents the premature
closure of a large number of small and often marginal well
operations. Not accounting for this size cutoff would similarly
contribute to differences in the estimated reduction in natural
gas production and employment losses associated with the proposed
standards.
Also, the coxmnenter has misinterpreted the EPA's
determination of closure, or abandonment, as based on aggregate
changes in corporate revenues and profits. Rather, as described
in Section 4 of the EIA report, the EPA's economic model
determines production and closure decisions on the basis of a
producing field (i.e., a group of similar wells) that is
consistent with the commenter's statement that "production
decisions are made on a well-by-well or project basis and if an
individual project's profits fall below its break-even point,
that the well will be abandoned." Furthermore, the commenter is
correct in stating that the EPA did not estimate losses of
2-112
-------
economically producible natural gas reserves. The economic
analysis conducted by the Agency is unable to address possible
impacts on production from future natural gas reserves. However,
based on the negligible impact on current natural gas production
associated with the EPA's engineering estimate of compliance
cost, it is not expected that these impacts would be as great as
indicated by the commenter.
2-113
-------
2.7 LEGAL ISSUES [OTHER THAN ISSUES ASSOCIATED WITH THE EPA'S
INTERPRETATION OF SECTION 112(n)(4)(A) AND (B)]
Comment: Commenters IV-D-16, IV-D-22, IV-D-23, and IV-D-34
requested that the EPA clarify that subparts HH and HHH do not
apply to OCS sources. Commenter IV-D-16 specifically recommended
that the definition be modified to clarify that offshore
platforms are not covered. The commenters indicated that
offshore platforms should not be covered by any section 112 rule.
Commenter IV-D-22 stated that they believe that Congress gave the
EPA limited authority to regulate air emissions from OCS sources.
According to the commenter, most of the authority for controlling
these emissions was provided to the Department of the Interior
(DOI) and what limited authority was provided to the EPA extends
only to emissions of criteria pollutants. Commenter IV-D-23 also
indicated that the CAA prevents the EPA from regulating HAP in
Federal OCS areas. Commenter IV-D-34 referred to section 328 of
the CAA and stated that it does not provide authority to the EPA
to regulate air toxics under section 112 in any OCS area.
Response: Section 328 of the CAA requires that the
Administrator establish requirements to control air pollution
from OCS sources (i.e., sources located offshore of the States
along the Pacific, Arctic, and Atlantic coasts and along the U.S.
Gulf Coast off the State of Florida eastward of longitude 87
degrees and 30 minutes) to attain and maintain Federal and State
ambient air quality standards and to comply with the provisions
of part C of title I. For sources located within 25 miles of the
seaward boundary of such States, the requirements must be the
same as would be applicable if the source were located in the
corresponding onshore area. Oil and natural gas production
sources emit VOC, which contributes to the formation of ozone,
and is regulated by the national ambient air quality standards
(NAAQS). The primary HAP of concern for these source categories
(benzene, toluene, ethyl benzene, mixed xylenes, and n-hexane)
2-114
-------
are also classified as VOC. Therefore, standards for oil and
natural gas production sources are applicable to offshore
platforms, that are OCS sources, because they are related to the
"attainment and maintenance" of ambient air quality standards or
the requirements of part C of title I of the Act. Furthermore,
section 328 states that the Administrator may exempt an OCS
source from a specific requirement "if the Administrator finds
that compliance with a pollution control technology requirement
is technically infeasible or will cause an unreasonable threat to
health and safety." Since offshore platforms typically control
process vents by routing them to a flare, the EPA has determined
that compliance with the control technology requirements is
technically feasible. Therefore, the EPA has not exempted
offshore platforms that are OCS sources from subpart HH.
2-115
-------
2.8 PERMIT ISSUES
Comment: Commenter IV-D-05 cited a problem with the
aggregation of emissions from associated equipment for major
source determinations. [Note: the commenter mentioned
"ancillary equipment" but most likely meant "associated
equipment" as the comment appears to be directed towards
associated equipment.] The commenter was concerned that
existing facilities that had made applicability determinations in
the past (e.g., for title V operating permits), based on not
aggregating emissions from glycol units and storage vessels with
the potential for flash emissions, had been permitted and
operated as minor sources. The commenter asked whether these
facilities would be given a "grace period" to pursue title V
operating permits if they were classified as major sources given
the aggregation of the ancillary equipment. The commenter was
concerned that without a grace period, these sites could be
subject to enforcement/penalty, etc.
Response; When making past applicability determinations,
sources may have interpreted the phrase "associated equipment" in
section 112(n)(4) of the Act differently than EPA'a final
interpretation of that phrase. The EPA acknowledges that such
sources may have concluded that they were nonmajor, whereas,
under EPA'a final subpart HH rule, they could be classified as
major. However, the EPA expects the number of sources with this
discrepancy is small. The majority of facilities that are major
under the EPA's final rule would have applied for title V permits
because they have emission points (e.g., glycol dehydration
units) that are by themselves major. Of the remaining sources,
the EPA expects many to have applied for a title V permit based
on the anticipated interpretation of 112(n)(4) described when the
proposed rule was published.
2-116
-------
For the remaining sources (e.g., those which are major
solely because of aggregation of associated equipment, and which
have not yet submitted a Title V permit application), the EPA
does not agree that a blanket policy granting a "grace period" is
appropriate, and EPA encourages major sources subject to subpart
HH to apply for a title V permit as expeditiously as possible.
The EPA will rely on its enforcement discretion in situations
where a source failed to apply for a permit because it determined
that it was nonmajor based on section 112(n)(4) of the Act. In
most cases the EPA does not expect to undertake enforcement
action, so long as the source expeditiously applies for its
title V permit. However, the EPA does not believe it is
appropriate to give up its ability to enforce part 70 in such
instances.
Comment; Commenter IV-D-06 requested that §63.1274(c)
mention that sources exempted by §63.1274(b) are not required by
this subpart to obtain an operating permit. According to the
commenter, sources exempted by §63.1274(b) should not be required
to get an operating permit, since subpart HHH has no requirements
to put into a permit. The commenter stated that this comment may
also apply to subpart HH.
Response; Under proposed §63.1274(b) [now being codified at
§63.1274(d)]/ only individual units are exempt from the
requirements of proposed §63.1274(a) [now being codified at
§63.1274(c)]. Therefore, major sources would still be required
to obtain a title V permit and include the part 63 requirements
that these sources keep records to document that the design
capacity or benzene emission rate is below the cutoff
(§63.1284(d) for subpart HHH and §63.764 for subpart HH) . The
EPA believes that, when a source is required to obtain a title V
permit because it is major, recordkeeping requirements like
2-117
-------
§63.1284(d) must be included. Note that in the final rules,
neither subpart HH nor subpart HHH regulates nonmajor sources.
Comment: Commenters IV-D-31 and IV-G-02 requested that the
sections that require major sources of HAP subject to the
proposal to get operating permits [§§63.764(f) and 63.1274(c)],
should be eliminated since the requirements are already well
documented in parts 70 and 71. The commenters explained that
restating these requirements is redundant and can cause confusion
in identifying applicable requirements in an operating permit
application.
Response; The EPA believes that stating the requirement for
a major source to obtain a part 70 or part 71 operating permit
identifies the facility's obligation to obtain such permit. The
EPA does not see any reason to remove §§63.764 (f) and proposed
63.1274(c) [now codified at §63.1274(e)].
Comment: Commenters IV-D-13 and IV-D-37 requested that the
NESHAP not affect the monitoring, recordkeeping, and reporting
requirements for control devices contained in the title V permit,
or other appropriate federal mechanism, at the time the final
NESHAP is promulgated. The commenters remarked that if the
control device is federally enforceable, the monitoring,
recordkeeping, and reporting requirements that have been
acknowledged as federally enforceable and quantifiable by the EPA
are sufficient to ensure HAP emission reductions.
Response; The final rules impose monitoring, recordkeeping
and reporting requirements that are independent from monitoring,
recordkeeping, and reporting for any other applicable
requirement. The EPA cannot assume that any existing control
device requirement, (including monitoring, recordkeeping, or
reporting) is adequate to ensure compliance with the particular
requirements of a new NESHAP. Ensuring compliance with the NESHAP
does not alter existing compliance obligations that are
2-118
-------
established for a variety of other reasons. However, the EPA
notes that adding the NESHAP to the title V permit may offer
opportunities to consolidate and streamline these multiple
applicable requirements if they exist. For additional discussion
on how this streamlining can occur, see the March 5, 1996 "White
Paper Number 2 for Improved Implementation of the Part 70
Operating Permits Program."
Comment: Commenter IV-D-14 asked what the role of State and
federal permit limits would be on the major source
determinations.
Response; Major source determinations are in part based on
a source's PTE. For the purposes of section 112, PTE is defined
in §63.2 such that any physical or operational limitation on the
capacity of a source to emit a pollutant, including air pollution
control equipment and restrictions on hours of operation, shall
be treated as part of the unit's design if the restriction is
federally enforceable. For additional information on limiting
PTE for section 112 purposes and for other reasons, please see
the following memoranda: (1) January 25, 1995 Memorandum from
John Seitz, Director, OAQPS, entitled "Options for Limiting the
Potential to Emit (PTE) of a Stationary Source Under Section 112
and Title V of the Clean Air Act;" (2) August 27, 1996 Memorandum
from John Seitz, Director, OAQPS, entitled "Extension of January
25, 1995 Potential to Emit Transition Policy;" and (3) July 10,
1998 Memorandum from John Seitz, Director, OAQPS, entitled
"Second Extension of January 25, 1995 Potential to Emit
Transition Policy and Clarification of Interim Policy."
2-119
-------
2.9 ENFORCEMENT ISSUES
Comment: Commenters IV-D-06, IV-D-07, and IV-D-14 stated
that parameter monitoring data do not, by themselves, demonstrate
compliance or noncompliance with emission standards. According to
these commenters, along with commenter IV-G-09, inability to
demonstrate compliance does not prove noncompliance. Commenter
IV-D-14 stated that the assumption that an emission limit has
been exceeded is not valid and the burden of proof for violation
of emission standards should lie on the agency enforcing the
rule.
Commenter IV-D-06 referred to the history of the HON and the
compromise between industry and the EPA on classifying monitoring
excursions as violations. According to the commenter, excursions
should not be classified as violations of the emission limit.
Instead, they should be classified as violations of an operating
requirement (the requirement to keep daily averages within the
approved limit). This commenter, along with commenters IV-D-12
and IV-G-09, suggested that is incorrect to define an excursion
as a violation of the emission standards. Commenter IV-D-14
asked what the basis was for establishing that a violation of an
operating parameter value automatically constitutes a violation
of an applicable emission standard.
According to commenter IV-D-06, subpart HHH should be
revised to require that industry "operate with the daily averages
within the approved limit, except as otherwise provided in this
subpart." The commenter further stated that subpart HHH should
then require that "excursions violate that paragraph." The
commenter requested that the EPA delete all portions of subpart
HHH that currently say excursions violate the emission standard.
According to the commenter, the EPA can assess the same
penalties. The commenter provided the following specific
portions of subpart HHH to be amended (Note: the commenter stated
that corresponding portions of subpart HH should also be amended,
if applicable).
• Section 63.1271 (definition of "operating parameter
value"): instead of "determines that an owner or
2-120
-------
operator has complied with an applicable emission
limitation or standard," say "indicates proper
operation of the control device."
• Section 63.1274(d): Delete.
• Section 63.1281(d) (4) (iii) ; Revise to say that,
except as otherwise provided in this subpart, any
excursion is a violation of the provisions of section
63.1281(d)(4)(ii).
In addition, commenters IV-D-07 and IV-D-12 recommended that
excursions outside the defined operating window should be a
notice for corrective or preventive action, instead of a
violation of the standard. According to commenter IV-D-12,
short-term excursions from the operating window do not result in
exceedence of "properly structured emissions limitations."
Commenter IV-D-07 claimed that the proposal is inconsistent with
other regulatory initiatives, such as the Compliance Assurance
Monitoring (CAM) rule.
In contrast, commenter IV-D-35 supported provisions, like
those contained in §63.764(h)(2), that plainly state that
noncompliance with operating parameters is a violation of the
emission limitation or standard.
Response; The EPA's decision to classify a violation of an
operating parameter as a violation of the emission standard
versus a violation of the operating requirement is based on
whether the monitored parameters have a strong correlation to
control device operation. In other words, do the monitored
parameters accurately predict control device performance? For
combustion units to achieve complete combustion, sufficient
reactor space, residence time, turbulence, and temperature are
necessary. A high combustion temperature must be provided to
ignite the vent stream HAP constituents. Therefore, since
reactor space, residence time, and turbulence are design
parameters, temperature can be used as an accurate prediction of
combustion device operation.
2-121
-------
For condensers, GRI has published a report entitled "Control
Device Monitoring of Glycol Dehydrators: Condenser Efficiency
Measurements and Modeling,"15 in which condenser outlet
temperature was evaluated as a sufficient monitoring parameter
for glycol dehydrator vent condensers. In the report, GRI
concluded that outlet temperature is a sufficient monitoring
parameter for indicating control device performance.
Because of these correlations for combustion devices and
condensers, the EPA believes that monitoring temperature is
strong indication of control device performance. The EPA
maintains that a violation of an operating parameter value should
be classified as a violation of the emission standard.
Therefore, the EPA has not made any changes to subparts HH and
HHH in response to this comment.
Comment: Commenter IV-D-06 stated that subpart HHH does not
appear to define what constitutes an excursion, although it
provides that excursions are violations. The commenter suggested
that excursions should be defined by the daily average parameter
value, not each monitored data point. The commenter recommended
that the EPA modify §63.1281(d)(4)(ii) to clarify that industry
is required to keep the daily average parameter value within the
limit. The commenter also recommended that the EPA include the
definition of an excursion shortly after this paragraph. The
commenter stated that this comment also applies to subpart HH.
Commenter IV-D-06 was also concerned that each individual
data point might be considered a violation of subpart HHH. The
commenter stated that a single missing data point is not an
excursion. Instead, a "data quality" excursion should mean that
15Reuter, C.O., et al (Radian International LLC). Control
Device Monitoring of Glycol Dehydrators: Condenser Efficiency
Measurements and Modeling, Volume 1. Prepared for the Gas
Research Institute. Publication Number GRI-97/0005.1 January
1997. 134 pp.
2-122
-------
less than 75 percent of the required data were collected. During
development of the HON, the industry stated that two types of
excursions exist: "parameter" and "data quality." The industry
contended that no matter how carefully the monitoring systems are
operated, sometimes a data point would not be recorded. The
industry stated that if 100 percent of the data were required to
be collected, compliance would be impossible. Commenter IV-D-35
was also concerned that the proposed regulations do not contain
many quality assurance/quality control provisions or a minimum
availability of time for the monitoring equipment. The commenter
suggested including a provision requiring 95 percent data
availability of continuous monitoring systems on an annual
(8,760 hours) basis. The commenters stated that this comment
also applies to subpart HH.
Response; The EPA agrees with the concepts suggested by the
commenters and has made several changes to subparts HH and HHH in
response with these comments. Section 63.773(d)(4) of final
subpart HH and §63.1283(d)(4) of final subpart HHH require the
owner or operator to calculate the daily average for each
monitored parameter and require that the daily average consist of
valid data points for at least 75 percent of the operating hours
in an operating day. For condensers, the owner or operator has
the option of converting the daily average temperature to an
annual average (subpart HH) or a 30-day average (subpart HHH)
condenser removal efficiency. In addition, the following
requirements have been added to §§63.773(d) and 63.1283(d):
1. An excursion for a given control device has occurred when
monitoring data or lack of monitoring data result in one of
the following:
• The daily average value of a monitored parameter is
less than the minimum operating parameter limit (or
greater than the maximum operating parameter limit, if
applicable) established for that operating parameter.
• If applicable, the 365-day average condenser efficiency
is less than 95 percent, unless the owner or operator
2-123
-------
has less than 365 days of data, the average condenser
efficiency is less than 90 percent.
• Monitoring data are not available for at least
75 percent of the operating hours.
• The vent stream has been diverted through the bypass
device.
2. Each excursion is a violation of the operating parameter
limit and thus a violation of the standard (either subpart
HH or HHH).
3. For each control device (or combination of control devices
installed on the same HAP emissions unit) , one excused
excursion is allowed for each semi-annual period
(corresponding to the periodic reporting periods specified
in §§63.775 and 63.1285).
4. Excursions are not considered violations and do not count as
excused excursions during the startup, shutdown and
malfunction events (provided the facility operates according
to the startup, shutdown, or malfunction plan) , and during
periods of nonoperation of the unit or process that is
vented to the control device.
gomment: Commenter IV-D-06 noted that subpart HHH allows
for owner or operator to choose the parameter limits and in some
cases may not be restricted to performance tests. The commenter
recommended that the EPA use a concept from the HON where the
owner or operator establishes the approved parameter limit, which
does not necessarily have to be based on a performance test. In
cases where the owner or operator is not otherwise required to
conduct a performance test, the parameter limit may be based on
engineering assessments or manufacturers' data if desired. The
commenter suggested that the EPA clarify subpart HHH by borrowing
from other MACT standards that explain in greater detail when to
use performance test data, and when using other data is
permissible, in establishing parameter limits. The commenter
suggested that the EPA use the Group I and Group IV Polymers
rules as a model. The commenter stated that this comment also
applies to subpart HH.
Response; The EPA does not intend to restrict when an owner
or operator should use performance test data or design analyses.
The owner or operator may decide whether a performance test or a
2-124
-------
design analysis will provide accurate data for determining an
appropriate minimum or maximum operating parameter value.
Proposed §§63.771(d)(3)(iii)(B) and 63.1281(d)(3)(iii)(B) [now
codified at §§63.772 (e) (4) (ii) and 63.1282(d)(4) (ii)] state that
if the owner or operator and the Administrator do not agree on
the demonstration of control device performance using a design
analysis, then the disagreement shall be resolved using the
results of a performance test.
Sections 63.773(d)(5) and 63.1283(d)(5) of the final rules
contain the requirements for establishing minimum (or maximum)
operating parameter limits. These requirements specify that the
owner or operator must establish the appropriate operating
parameter limit using performance test data, or a design
analysis. Both performance test data and the design analysis may
be supplemented using control device manufacturer's information.
Comment: Commenter IV-D-06 noted that subpart HHH does not
allow for any excused excursions. The commenter recommended that
each monitored, control device or recovery device be given a
specified number of excused excursions where the number of
excused excursions starts larger and becomes smaller with time.
The commenter referred to the development of the HON, where the
industry raised the concern that no matter how carefully a
control device is operated and maintained, sometimes there will
be an excursion. The industry maintained that these excursions
are most frequent when a control device is new and is being
debugged, but they will decrease over time. Therefore, the
commenter recommended that the EPA modify §63.1281(d)(4)(ii) by
using §63.152(c) (2) (ii) (B) of the HON, which says:
(B) The number of excused excursions for each
control device or recovery device for each semiannual
period is specified in paragraphs (c) (2) (ii) (B) (1)
through (c)(2)(ii)(B)(£) of this section. This
paragraph applies to sources required to submit
Periodic Reports semiannually or quarterly. The first
2-125
-------
semiannual period is the 6-month period starting the
date the Notification of Compliance Status is due.
(1) For the first semiannual period - six excused
excursions.
(2.) For the second semiannual period - five
excused excursions.
(3_) For the third semiannual period - four
excused excursions.
(4.) For the fourth semiannual period - three
excused excursions.
(£) For the fifth semiannual period - two excused
excursions.
(£) For the sixth and all subsequent semiannual
periods - one excused excursion.
The commenter stated that this comment also applies to
subpart HH.
Response; The compliance dates for subparts HH and HHH
allow owners and operators three years after the effective date
of the rule to achieve compliance. The EPA believes that there
is sufficient time for an owner or operator to debug control
devices and monitors. Furthermore, by allowing for a small
amount of missing data, and by specifying that a violation of the
operating parameter is defined by the daily average parameter
value, the EPA believes that owners and operators have sufficient
flexibility to operate and maintain their control devices.
Therefore, the EPA maintains that only one "excused excursion"
should be allowed per semiannual period [codified at
§§63.773(d) (8) and 63 .1283 (d) (8) ].
Sections 63.773(d)(8) and 63.1283(d)(8) also state that
during startup, shutdown and malfunction events, as long as the
owner or operator complies with the facility's startup, shutdown
and malfunction plan, any monitored parameters outside its
operating range would not be counted towards the excused
excursions. However, simply following a startup, shutdown, and
malfunction plan is not necessarily a defense to failure to have
taken steps to prevent malfunctions or failure to adequately
2-126
-------
minimize emissions during startup, shutdown, and malfunction
events (§§63.762 and 63.1272). Also, during periods of
nonoperation of the unit or process that is vented to the control
device, monitored parameters outside their established operating
ranges do not count as excursions.
Comment: Commenter IV-D-06 stated that monitoring data
collected during startups, shutdowns, and malfunctions, or
periods of non-operation, should be excluded from daily averages.
The commenter noted that since daily averages are not mentioned,
this concept is also not mentioned in the subpart HHH. The
commenter stated that in the General Provisions, normal emissions
standards do not apply during startups, shutdowns, and
malfunctions. During those periods, compliance is determined
based on the facility following the provisions in the startup,
shutdown, and malfunction plan. The commenter recommended that
the EPA borrow the following concepts from §§63.152(c) (2) (ii) (C)
and 63.152 (f) (7) of subpart G and incorporate them into
§63.1281(d)(4)(ii) (note that wording would have to be changed):
(C) If a monitored parameter is outside its
established range or monitoring data are not collected
during periods of startup, shutdown, or malfunction
(and the source is operated during such periods in
accordance with the source's startup, shutdown, or
malfunction plan as required by §63.6 (e) (3) of subpart
A of this part) or during period of non-operation of
the chemical manufacturing process unit or portion
thereof (resulting in cessation of the emissions to
which the monitoring applies), then the excursion is
not a violation and, in cases where continuous
monitoring is required, the excursion does not count
toward the number of excused excursions for determining
compliance.
• • •
(7) Monitoring data recorded during periods
identified in paragraphs (f)(7)(i) through (f)(7)(v) of
this section shall not be included in any average
computed under this subpart. Records shall be kept of
the times and durations of all such periods and any
other periods during process or control device
operation when monitors are not operating.
(i) Monitoring system breakdowns, repairs,
calibration checks, and zero (low-level) and high-level
adjustments;
2-127
-------
(ii) Startups;
(iii) Shutdowns;
(iv) Malfunctions; and
(v) Periods of non-operation of the chemical
manufacturing process unit (or portion thereof),
resulting in cessation of the emissions to which the
monitoring applies.
The commenter stated that this comment also applies to
subpart HH.
Response; As stated in a previous comment, the EPA has
included provisions for one excused excursion and has
incorporated the suggested concepts from §63.152(e)(2)(ii) (C)
[codified at §63.773(d)(8) and 63.1283(d)(8) of the final rules].
The EPA also agrees that monitoring data collected during
startups, shutdowns, and malfunctions, or periods of
non-operation, should be excluded from the daily averages.
Therefore, the EPA has incorporated the suggested concepts from
63.152 (f) (7) of subpart G and incorporated them into
§§63.774(b) (3) and 63.1284(b)(3) of the final rules.
Comment: Commenter IV-D-06 expressed concern that each
monitored parameter would be considered a separate violation of
subpart HHH. According to the commenter, if a control device has
two or more parameters that must be monitored, and more than one
parameter has a daily average outside the approved limit on the
same day, this should be considered a single excursion. During
the development of the HON, the industry explained that operating
parameters are generally interrelated, so no matter how many
parameters are outside the limit, there is only one opportunity
for emissions to be above the standard. The industry felt that
it would be unfair to multiply the violations by considering each
parameter separately.
Response; The EPA agrees with the commenter. Sections
63.773(d)(6) and 63.1283(d)(6) of the final rules state that for
a control device or recovery device where multiple parameters are
monitored, if one or more of the parameters meets the criteria
2-128
-------
for an excursion, this is considered a single excursion for the
control device.
Comment: Commenter IV-D-14 asked how the EPA would
initially administer the new program. The commenter also asked
what the States' role would be in administering the program and
how delegation would be afforded to the States.
Response; Section 112(1) of the Clean Air Act grants the
Administrator the authority to approve State programs to
implement and enforce section 112 rules. Subpart E of part 63
establishes the procedures for States to follow in obtaining
delegated authority as provided in section 112 (1). This subpart
establishes the procedures for:
• the approval of State rules or programs to be
implemented and enforced in place of.section 112
Federal rules, emission standards, or requirements;
• the approval of State programs to implement and enforce
section 112 Federal rules as promulgated without
changes; and
• the approval of State rules or programs that adjust a
section 112 Federal rule.
Any request for approval under subpart E must meet all
section 112(1) approval criteria specified by the applicable
Federal rule, and the approval criteria in §63.91(b) of
subpart E. The EPA expects that by the compliance dates of
subparts HH and HHH, the States programs to implement and enforce
these subparts will have been approved by the Administrator under
subpart E. Delegation of authority will be specified in §§63.776
and 63.1286. However, in the case that delegation is not made,
then the EPA Regional Administrator for that State would
implement the standard.
Comment.- Commenters IV-D-07, IV-G-09, and IV-D-31 were
concerned with the limitations of GLYCalc and noted that the
GLYCalc instruction manual states that it over predicts
2-129
-------
emissions, usually by at least 20 percent. The commenters
recommended GLYCalc not be used for enforcement purposes unless a
disclaimer on its use for enforcement purposes is included.
Response; The EPA performed field tests to assess the
effectiveness of the GLYCalc emissions model for estimating HAP
and VOC emissions.16 Based on the results of the field test
evaluations, and additional glycol dehydrator emissions test
sponsored by 6RI and API, the EPA has recommended that the
GLYCalc model be included in guidance for State and local agency
use for the development of emissions inventories to meet CAA
requirements. The EPA stated that for sites where source tests
have been conducted, the experience was that GLYCalc either
estimates emissions accurately or overestimates emissions.
According to the EPA's analysis, the likelihood of overestimating
emissions may be reduced by obtaining accurate measurements of
process variables for as many model inputs as possible. However,
since the use of default values for model inputs will
occasionally be necessary, some overestimation of emissions is
unavoidable.
Therefore, based on the EPA's analysis, the EPA believes
that GLYCalc is a reasonable method for estimating benzene
emissions from glycol dehydration units for the 1-tpy exemption
and for use in conjunction with the Atmospheric Rich/Lean (ARL)
method as an alternative to the performance procedures for
condensers.
It should be noted that the EPA does not require the use of
GLYCalc, but has offered it as an acceptable tool for estimating
"Memorandum from Jones, L.G., U.S. EPA/Emissions
Measurement Branch, to J.D. Mobley, U.S. EPA/ Emission Factor and
Inventory Group. "Glycol Dehydrator Emissions Test Report and
Emissions Estimation Methodology." April 13, 1995.
2-130
-------
emissions and demonstrating compliance. However, owners and
operators should be aware that it is possible that a performance
test could indicate that a glycol dehydration unit is out of
compliance. Therefore, the EPA recommends that if GLYCalc
predicts that the glycol dehydration unit is operating close to
the emission limitations in the NESHAP, then the owner or
operator may wish to conduct a performance test.
2-131
-------
2.10 CONTROLS
2.10.1 MACT Floor
Comment: Commenter IV-D-07 noted that the EPA assumed an
average inlet BTEX concentration of 200 ppmv as the basis for the
95 percent HAP reduction. According to the commenter, if the
BTEX concentration is well below 200 ppmv, the required reduction
would be much more difficult to obtain for condensers. The
commenter stated that they felt the EPA did not consider this
scenario or allow for a cost-effective solution. The commenter
noted that for a combustion device, the 95 percent reduction
should not be a problem.
Response; The commenter' s statement that the 95 percent HAP
reduction requirement was based on an average inlet BTEX
concentration of 200 ppmv is incorrect. Instead, the 95 percent
control requirement was developed as the floor level of control
(see Air Docket A-94-04, number II-A-07 for further discussion on
the MACT floor development).
The national average BTEX concentration was developed to
estimate national emissions. The EPA developed three national
average BTEX concentrations for natural gas to represent three
sectors in the oil and natural gas production and natural gas
transmission and storage source categories: (1) 200 ppmv for
production; (2) 160 ppmv for processing; and (3) 13 ppmv for
transmission and storage.17 To develop the national emission
impacts, the EPA distributed concentrations of BTEX among the TEG
unit populations.18 Therefore, since the 95-percent emission
reduction was not based on the average BTEX concentrations, the
EPA did not see any reason to modify subpart HH in response to
this comment.
17Reference 9, appendix B.
18Reference 9. Appendix B.
2-132
-------
However, the final rule requires an owner or operator to
control process vents on glycol dehydration units to one of the
following: (1) 95 percent HAP emission reduction, (2) 20 ppxnv
control device outlet concentration (for combustion devices), or
(3) control device outlet mass emission rate less than 1 tpy of
benzene. The benzene limitation was established because the MACT
floor for glycol dehydration units with actual benzene emissions
less than 1 tpy was determined to be no control (see responses in
section 2.10.3 of this document). The EPA believes that the
addition of the 1 tpy benzene emission limitation provides
additional flexibility for owners or operators of facilities with
low BTEX concentrations in the natural gas.
Comment: Commenter IV-D-06 requested that the EPA allow an
emission limit of 20 ppmv for non-combustion" control devices.
The commenter stated that at very low incoming HAP
concentrations, recovery devices may be unable to achieve
95 percent HAP reduction, but can probably achieve 20 ppmv HAP at
the outlet reliably. The commenter was concerned that they would
be forced to use combustion devices rather than recovery devices.
The commenter remarked that the combustion device would be
allowed to emit the same 20 ppmv that the recovery device was not
allowed to emit and that recovery for reuse is environmentally
better than destruction. The commenter recommended that the EPA
add a 20 ppmv option to §63.1281(d) (1) (ii), using
§63.1281(d)(1)(i)(B) as a pattern, without the correction to
3 percent oxygen. The commenter stated that this comment also
applies to subpart HH.
Response; In the preamble to the proposed 40 CFR part 60,
subpart NNN, NSPS for Air Oxidation Unit Process (48 PR 48932,
October 21, 1983), the EPA stated that 20 ppmv is the lowest
outlet concentration of total organic compounds achievable by the
combustion of low organic concentrations (i.e., inlet
2-133
-------
concentrations of 2000 ppmv or less). As stated in the preamble
to subpart NNN, the outlet concentration was established based on
kinetic calculations of incinerators. It was demonstrated that,
at a given temperature and residence time, a stream with a low
inlet concentration (approximately 2000 ppmv) could not be
controlled in an incinerator to an outlet concentration below
20 ppmv. The coxnmenter did not provide any information
indicating that non-combustion control devices could not meet an
outlet concentration below 20 ppmv. Therefore, the EPA does not
see any reason to modify subparts HH and HHH in response to this
comment.
Comment;: Commenter IV-D-21 was concerned that documenting
95 percent reduction might be difficult for some dehydrator
configurations that have a flash tank. According to the
commenter, less than 2 percent of the total uncontrolled HAP
emissions from a dehydrator are associated with the flash gas.
The commenter explained that for dehydrators that route vent gas
to a condenser to recover hydrocarbons, and route flash gas to a
combustion device, the compliance determination would depend on
defining the emission reduction achieved by both control devices.
Based on the small amount of HAP emissions associated with the
flash gas, the commenter suggested that testing the combustion
device in accordance with proposed §63.772 (e) to document control
efficiency would significantly increase compliance cost with
little environmental benefit.
The commenter also noted that flash gas has a high British
thermal unit (Btu) content and is easily burned. The commenter
requested that the EPA provide a default reduction efficiency in
the final rule that can be used by combustion systems burning
flash gas for use in demonstrating compliance with the
requirement for a 95-percent reduction efficiency.
Response: The EPA believes that subparts HH and HHH provide
sufficient flexibility in demonstrating compliance for owners and
2-134
-------
operators that route the glycol dehydration unit reboiler vent
gases to a condenser for hydrocarbon recovery and route the flash
gas vent to a combustion device. First, final subparts HH and
HHH allow combinations of control devices to achieve the
95 percent emission reduction [final §§63.765(b)(1)(i) and (ii)
and 63.1275(b) (1) (i) and (ii)]. Second, the final rules do not
require control of HAP emissions from flash tanks if the total
HAP emissions to the atmosphere from the glycol dehydration unit
process vent (i.e., the combined reboiler and flash tank vents)
are reduced by 95 percent [§§63.765(c) (3) and 63.1275(c) (3)] .
Finally, subparts HH and HHH provide owners and operators the
option of demonstrating compliance using either a performance
test or a design analysis [§§63.772(e) and 63.1282(d)]. The EPA
believes that by providing the option of performing a design
analysis rather than a performance test, owners and operators
have the flexibility to choose the least expensive option.
As for the commenter's request for a default reduction
efficiency for combustion devices burning flash gas, the EPA
points to §63.771(d)(1)(i) in final subpart HH and
§63.1281(d)(1)(i) in final subpart HHH which state that an owner
or operator may install enclosed combustion devices that meet one
of the following conditions: (1) reduces HAP emissions by
95 percent or more, (2) reduces the outlet HAP concentration to
20 ppmv or less, (3) operates at a minimum residence time of 0.5
seconds at a minimum temperature of 760°C, or (4) is boiler or
process heater that is designed so the vent stream is introduced
into the flame zone. If the owner or operator can demonstrate
that their combustion device operates according to the minimum
residence time and temperature specifications or the vent stream
is introduced into the flame zone, then a compliance
2-135
-------
demonstration with the 95-percent emission reduction (or the
20 ppmv outlet concentration) is not required. Therefore, the
EPA believes that it is not necessary to provide a default
reduction efficiency for demonstrating compliance with the
95-percent reduction efficiency.
Comment: Several commenters objected to the 95-percent
control requirement. According to commenter IV-D-12, a
90 percent control requirement would provide a more realistically
achievable standard. The commenter pointed to a GRI study
(Skinner and Rueter, 1998, GRI-98/0073) which shows that
condensers at many facilities are unable to achieve 95 percent on
a continuous basis. The commenter further stated that applying
the control requirement based on the source's choice of either
total VOC or HAP is appropriate, but not both.
Commenter IV-D-16 stated that condenser performance is
dependent on local climate conditions, and that a 95-percent
efficiency cannot be reliably achieved throughout the year in
many areas of the United States. The commenter was concerned
that requiring efficiencies that could not be met would remove
condensers from the potential control list. The commenter
further remarked that condensers are a form of recycling and
should not be "saddled with an efficiency requirement they cannot
meet." The commenter suggested that an efficiency of 85 percent
more accurately describes the performance in these devices and
should be chosen as the required level of control for subparts HH
and HHH. Commenter IV-D-38 also recommended that an emission
reduction of 85 percent would be typically achieved for the
control devices addressed in section 63.765(c)(2) and (3), as
compared with 95 percent in the proposed regulations, and that
the cited paragraphs should be changed accordingly.
Commenter IV-G-07 presented HAP efficiency data for twenty
condenser-controlled glycol units, each treating between 5 and
55 MMscf/d, and only one of which has no flash tank. According
to the commenter, the calculated mean annual control efficiency
based on this data is 95.97 percent and the standard deviation
2-136
-------
error is 2.81 percent. Therefore, the commenter recommended 90
percent as the appropriate lower limit, using this data as
representative and taking the common scientific approach of using
plus or minus two standard deviations as the proper confidence
limit.
Commenter IV-G-02 was concerned about the EPA's setting a
MACT floor for dehydration units based on a "control level
estimated to be achieved through the use of condensers" (63 FR
6304), rather than on the "emission limitation achieved" as
required by the Act. The commenter questioned whether the EPA
has considered that many condensers located at an altitude
probably cannot meet 95 percent (due to lower atmospheric
pressure, substances exhibit a greater partial pressure, and are
therefore more difficult to condense). The commenter also
questioned whether the EPA has data to show the average of the
best performing 12 percent of units achieve a 95 percent
reduction. The commenter stated that they believe that the MACT
floor is an equipment standard (rather than efficiency, which is
appropriate) requiring a condenser, combustion device, or flare
for control.
Response; The MACT standard for process vents on new and
existing glycol dehydration units was set at the floor level of
control. As required under section 112(d) of the Act, the EPA
developed the MACT floor based on ". . . the average emission
limitation achieved by the best performing 12 percent of the
existing sources. ..." A detailed discussion regarding the
development of the MACT floor can be found in the docket (Air
Docket A-94-04, number II-A-07). Through section 114
questionnaires, site visits, meetings with stakeholders, and
available literature, the EPA obtained information for 200 glycol
dehydration units that were considered to be major sources of HAP
(prior to control). Of these, 34 percent (67 units) were
controlled using a variety of control technologies, including:
2-137
-------
condensation, combustion, and a combination of condensation and
combustion. The types of control technologies used by the
industry have been demonstrated, in other applications, to
achieve varying levels of emission reduction (ranging from 95 to
98 percent or better). The EPA could not identify a technical
basis for the variation in the performance levels achieved by the
controls reported to be used to control process vents on glycol
dehydration units. In order to account for the variability in
HAP emission reduction efficiencies, the EPA selected
95.0 percent as the required emission reduction (i.e., the MACT
floor) for glycol dehydration units in the oil and natural gas
production source category. Since the 95-percent emission
reduction allows owners and operators to install not only
condensers, but also combustion devices, as long as they achieve
a 95-percent HAP emission reduction, the EPA does not believe
that the MACT floor is an equipment standard.
Although the EPA did not lower the required emission
reduction in the final rule, the final rule requires compliance
with the 95-percent HAP emission reduction to be demonstrated on
a daily basis with an option for compliance using condensers to
be demonstrated using a 365-day rolling average for the oil and
natural gas production source category, and a 30-day rolling
average for the natural gas transmission and storage source
category (see section 2.10.2 of this document for further
discussion on averaging periods).
Regarding the commenter's concern about condensers located
at a high altitude, although differences in altitude do affect
condenser performance, the EPA expects the effect to be minimal.
Comment: Commenter IV-D-01 requested that the EPA
reevaluate the MACT floor for new sources and require new sources
2-138
-------
to control HAP emissions by 98%. According to the commenter,
Louisiana has many facilities that achieve a 98% or greater
control device efficiency by means of a condenser and closed vent
system routing non-condensables to the glycol reboiler firebox.
Response; Based on the available information (i.e.,
primarily section 114 questionnaire responses), the EPA did not
identify a method of control applicable to all types of new
sources that would achieve a greater level of HAP emission
reduction than the MACT floor for existing sources. Furthermore,
the EPA believes that requiring 98 percent emission reduction for
sources in the oil and natural gas production and natural gas
transmission and storage source categories would involve the
destruction of nonrenewable resources and does not encourage
pollution prevention.
Comment: Commenter IV-D-25 emphasized that catalytic
incineration is an effective control option. According to the
commenter, minimum temperature and residence time requirements
are lower for catalytic oxidation as compared with thermal
oxidation, resulting in less expense required for fabrication.
The commenter requested that §63.771(d)(1)(i) be modified to
include: "(D) For catalytic incineration, operates at a minimum
residence time of 0.03 to 0.05 second at a minimum temperature of
340°C." The commenter stated that these requirements would be
adequate for more than 95 percent destruction of the HAP.
Response; The EPA believes that the 0.5 second residence
time and 760°C minimum temperature requirements for enclosed
combustion devices are sufficient to ensure compliance with the
95-percent HAP emission reduction requirement.19 The commenter
19 U.S. Environmental Protection Agency. Hazardous Air
Pollutant Emissions from Process Units in the Synthetic Organic
Chemical Manufacturing Industry - Background Information for
Proposed Standards. Volume IB: Control Technologies. EPA
(continued..
2-139
-------
did not provide any data to justify that a 0.03 to 0.05 second
minimum residence time and 340°C minimum temperature would be
adequate to achieve a 95-percent HAP emission reduction for all
catalytic incinerators. Furthermore, the EPA does not have the
available data to determine whether catalytic incinerators with
these minimum specifications would meet the 95-percent emission
reduction requirements. Therefore, the EPA has not modified
subparts HH and HHH in response to this comment.
However, owners or operators may use a catalytic incinerator
with the parameters specified by the commenter, provided the
performance test or design analysis [prepared as specified in
§§63.772(e) and 63.1282(d)] shows that the control device meets
the required HAP emission reduction efficiency.
2.10.2 Averaging Period
Comment: Commenters IV-D-08 and IV-D-20 noted that
§§63. 771 (d) (1) (i) (A) and 63.771 (d) (1) (ii> do not state averaging
periods for the 95 percent control efficiency determination.
Several commenters were concerned with demonstrating compliance
with the 95 percent control efficiency on a continuous basis.
Commenter IV-D-12 stated that the proposal for a 15-minute
averaging period is inappropriate and could not be consistently
achieved due to swings in ambient conditions over which the
source has no control. Commenters IV-D-04, IV-D-08, IV-D-12,
IV-D-15, IV-G-05, and IV-G-09 requested that the EPA require
calculation of control efficiency on a monthly or 30-day basis
and commenters IV-D-20, IV-D-22, IV-D-23, IV-D-30, IV-D-34,
IV-G-02, requested a 12-month rolling or annual basis for all TEG
units subject to a 95 percent control requirement. Commenter
IV-D-31 supported either a 30-day or a 12-month averaging period.
Commenter IV-G-02 stated that a 12-month rolling average is more
19 (. . .continued)
Number EPA-453/D-92-016b., November 1992.
2-140
-------
appropriate, considering the types of risks involved, still
provides the EPA with enforceable numbers, and more appropriately
reflects the frequency of the reporting periods required in the
proposal. Commenter IV-G-03 stated that continuous compliance
could be achieved in the winter months and recommended that
continuous compliance determination be based on an annual
average, using rolling monthly data. All of the commenters
maintained that using a longer averaging period would create no
significant change in the emissions to the environment, but would
substantially decrease the number of technical violations of the
standard, and reduce the administrative burden for the industry
and the EPA. Commenter IV-D-04 explained that flow conditions in
dehydrators fluctuate over time and a 15-minute compliance period
would cause many units to be out of compliance that would be in
compliance over a longer period. The commenter suggested that
the shorter averaging time would make the control requirement
more rigorous than the EPA may have intended. Moreover,
commenters IV-D-20 and IV-D-22 stated that they believe longer
averaging periods are consistent with the MACT floor.
Commenters IV-D-10, IV-D-20, IV-D-22, IV-D-30, IV-D-34, and
IV-G-11 stated that the data collected under section 114 do not
support a MACT floor determination of 95 percent on a continuous
basis. According to the commenters, the section 114
questionnaire did not ask for the averaging period. Furthermore,
commenters IV-D-20, IV-D-22, and IV-G-11 stated that respondents
to the EPA's section 114 survey most likely did not provide
estimated efficiency on a continuous basis because the data to
make that evaluation were not available. According to the
commenters, to provide an estimate of condenser efficiency, most
respondents would have relied on vendor data or short duration
tests and would not have considered seasonal or diurnal
variations.
In support of a "monthly or annual averaging period,
commenters IV-D-07, IV-D-08, IV-D-10, IV-D-15, IV-D-20, IV-D-22,
IV-D-23, IV-D-27, and IV-D-31 stated that condensers and flash
tanks cannot achieve 95-percent HAP reduction continuously during
2-141
-------
the hotter months. The commenters referenced a report by GRI,20
which illustrated that high ambient temperatures cause the
control efficiency to drop below 95 percent. However, the report
showed that condensers could meet 95-percent control using a
longer averaging period. Commenters IV-D-20, IV-D-22, IV-D-31,
and IV-G-11 noted that, in the report, three fourths of the TEG
units controlled by condensers and flash tanks do not achieve a
95-percent reduction on an hourly basis. Commenters IV-D-27 and
IV-D-31 recommended that the EPA review the GRI report and adjust
the control efficiency and averaging time as appropriate.
Based on the GRI study21, commenters IV-D-08, IV-D-10,
IV-D-15, IV-D-20, IV-D-22, IV-D-34, and IV-G-11 were concerned
that to achieve 95-percent control on a continuous basis,
additional combustion controls would be necessary. Commenter
IV-D-10 referred to the supplementary information in which the
EPA mentions that flares and other combustion devices were not
included in the MACT floor analysis because they do not recover
hydrocarbons. The commenter agreed that an after condenser
combustion device would waste nonrenewable resources for the sake
of peaks in ambient temperatures. Commenter IV-G-11 also noted
that many operators would install flares or incinerators rather
than reroute vapors to the firebox (due to safety issues and
State opacity regulations). Furthermore, commenters IV-D-10 and
IV-D-15 were concerned that the combustion device would force
operators to make tradeoffs among emissions of NOX, VOC, and HAP.
Commenter IV-D-10 stated that combustion devices increase
emissions of NOX and VOC with greater dispersion impacts for only
5 percent additional HAP control on warm days and that the MACT
floor is not achievable with the technology envisioned.
20GRI, "Investigation of Condenser Efficiency for HAP
Control from Glycol Dehydrator Reboiler Vent Streams: Analysis
of Data from the EPA 114 Questionnaire and GRI's Condenser
Monitoring Program," Table 3-1 (March 1998).
^Reference 20.
2-142
-------
Commenter IV-G-12 stated that they collaborated several
years ago in an investigation of an evaporatively (water) cooled
condenser at one of its facilities that showed the condenser
could capture a significant portion (>90%) of the volatile
fraction coming from the still vent of a dehydrator. According
to the commenter, these condensers are easy to operate, provide
significant control, and result in the recovery and conservation
of a useful hydrocarbon stream. The commenter was concerned that
the proposed rule's presumed short term requirement of 95-percent
efficiency will preclude the use of devices of this type. The
commenter stated that the makeup of the gas being processed
determined the type of control that can be used. The commenter
suggested that facilities that can use a condenser should be able
to do so at a lower efficiency than required for flares operating
on equipment where condensers are not viable.
Commenter IV-D-21 provided emission reduction data resulting
from tests on dehydrators equipped with R-BTEX condensers.
According to the commenter, the tests showed 96 to 98 percent VOC
and HAP emission reductions, at ambient wet bulb temperatures
ranging from 65 to 85°F. The commenter stated that while they
are confident that these condensers could achieve a 95-percent
control efficiency on an annual basis, they were concerned that
brief periods of high ambient temperatures would result in lower
control efficiencies. To account for the impact of ambient
temperature, the commenter requested that the EPA establish an
averaging time during which the average condenser outlet
temperature must comply with the requirements proposed in
§§63.773 and 63.1283.
Commenters IV-D-08 and IV-D-20 objected to the EPA's
suggestion that continuous compliance with a standard is
necessary to protect human health and the environment from
emissions from this source category. Commenter IV-D-20 urged the
EPA to retract this statement as unsupported and inconsistent
with other regulatory programs. The commenter suggested that the
industry is sensitive to unnecessary control costs, and urged the
EPA to reconsider this requirement. Commenters IV-D-08 and
2-143
-------
IV-D-20 stated that continuous monitoring as required in
§63.773(d) is not necessary because a monthly average compliance
demonstration does not increase emissions. Commenter IV-D-22
indicated that a rolling 12-month average compliance
demonstration would not result in significant increases in annual
emissions. Commenter IV-D-22 recommended that periodic
monitoring be substituted at an interval appropriate to the
condenser averaging period.
Commenter IV-G-07 stated that year-to-year variations in
annual temperature histograms are small and that using an annual
temperature histogram to calculate annual emissions is an
excellent proxy for any year. Furthermore, the commenter
maintained that there is not a good proxy for any day. The
commenter stated that Congress's focusing on annual emissions in
the Clean Air Act Amendments (CAAA) of 1990 set forth a good
regulatory policy and asked why should this policy be changed.
Commenters IV-D-08, IV-D-10, IV-D-15, IV-D-20, IV-D-22,
IV-D-34, IV-G-03, and IV-G-11 stated that for units with existing
condensers that do not quite achieve 95-percent reduction, the
incremental cost to remove a small increment of HAP emissions is
cost prohibitive. Commenters IV-D-20 and IV-D-22 stated that the
marginal cost to remove a small increment of HAP emissions to
achieve 95-percent control on a continuous basis would exceed
$ 20,000/ton of HAP removed. Commenters IV-D-34 and IV-G-11
estimated this cost to be $ 30,000/ton.
Response; Based on the information available to the Agency,
the EPA believes that the control devices required by the final
rule achieve 95-percent HAP emission reduction on a daily basis.
However, the EPA has reviewed the GRI reports regarding condenser
performance22 and has considered the commenters concerns
regarding averaging periods for condensers. Based on this
^Reference 20.
2-144
-------
information, the Agency has included an option for owners and
operators that install condensers.
Under the final subpart HH [§63.772(f)], an owner or
operator of a glycol dehydration unit subject to the control
requirements under final §63.765 must demonstrate compliance with
the control device performance requirements on a daily basis. As
an alternative, the owner or operator that uses condensers to
comply with the requirements of §63.765 has the option of
demonstrating compliance with the 95-percent HAP emission
reduction on a 365-day rolling average [§63.772(g)]. An owner or
operator with less than 120 days of condenser operating data is
not required to calculate the average condenser efficiency until
after the first 120 days of operation. If this average
efficiency is equal to or greater than 90 percent, the owner or
operator is in compliance. Owners or operators with 120 days or
more, but less than 365 days of condenser operating data, must
calculate the average condenser efficiency over the number of
days of operation between the current day and the applicable
compliance date [specified in §63.760 (f)]. The owner or operator
is considered to be in compliance with the performance
requirements if this average condenser efficiency is equal to or
greater than 90 percent. Once the owner or operator has 365 days
of condenser operating data, the owner or operator must comply
with the 95 percent HAP emission reduction requirement on a
365-day rolling average.
For glycol dehydration units in the natural gas transmission
and storage source category, the EPA believes that an averaging
period shorter than 365 days is appropriate. To the Agency's
knowledge, glycol dehydration units located at storage facilities
do not typically operate throughout the year. Additionally,
2-145
-------
glycol dehydration units located at these sources do not
typically operate during the warm summer months when condenser
efficiency is lower. The data for the GRI report was based on
the operation of production facilities. Although transmission
facilities do operate for most of the year, the EPA believes that
the HAP emission units in operation at these facilities are
primarily compressors and that most glycol dehydration units
located at transmission facilities are used for withdrawing
natural gas from storage (i.e., are not likely to operate year-
round) . Therefore, the final subpart HHH specifies that owners
or operators that install condensers have the option of complying
with the 95-percent HAP emission reduction on a 30-day rolling
average [§63.1282(f)]. However, §63.1282 (f)(2) (iii) (D) of final
subpart HHH provides the owner or operator with the option of
complying with the 365-day rolling average procedure specified in
§63.772(g) for glycol dehydration units in the natural gas
transmission and storage source category that are operated
continually.
2.10.3 Process Vent Standards
Comment: Commenter IV-D-08 stated that §63.771(d) should be
modified to provide credit for total reductions and cumulative
efficiency rather than requiring "control upon control"
efficiency. The commenter recommended changes to
§63.771(d)(1)(i), (ii), and (iii) to allow cumulative reductions
for control devices in series:
Reduce or contribute to the reduction of the mass
content of either Total Organic Compound (TOG) or total
HAP by 95 percent, from the point that gases are vented
to the first control device until the point that gases
are vented to the atmosphere.
Commenter IV-D-06 requested that subpart HHH expressly allow
combinations of control devices as a way to achieve the emissions
standards. According to the commenter, it may take two or more
2-146
-------
control devices to achieve the emission limit. For example, some
units are controlled by devices that may come close to, but do
not meet the requirements. The commenter stated that it is
sometimes quicker, easier, and less expensive to add a
supplemental device than to remove an existing device and install
another. The commenter suggested using HON §§63.113(a) (2) (i),
(ii), and (ii)(A), (B), (C), and (D) to address combinations of
control and/or recovery devices. The commenter stated that if
subpart HH does not expressly allow combinations of control
devices, this comment applies to subpart HH.
Response; The EPA agrees that owners or operators should be
able to comply with the requirements of §§63.765 and 63.1275
using combinations of control devices. Therefore, the EPA has
modified subparts HH and HHH to allow an owner or operator to
connect glycol dehydration unit process vents to a control device
or a combination of control devices [§§63.765(b)(1) and
63.1275(b) (1)] . In addition, the EPA has modified §63.772 (e)(3)
of final subpart HH and §63.1282(d)(3) of final subpart HHH to
require the sampling sites to be located at the inlet of the
first control device and at the outlet of the final control
device.
Comment: Commenters IV-D-06, IV-D-07, IV-D-08, IV-D-20,
IV-D-22, and IV-D-30 requested that the EPA allow any
combinations of controls and process modifications to achieve the
required control efficiency. Commenters IV-D-08, IV-D-20,
IV-D-22, and IV-D-30 recommended that the EPA modify
§63.765(c)(2) to add language specifically stating that process
modifications and controls are allowed.
In addition, commenter IV-D-30 suggested that the EPA
specify in §63.765(c) that the owner or operator may elect to
complete a compliance demonstration once for the required process
modifications. According to the commenter, no more demonstration
should be necessary if the owner or operator made only process
2-147
-------
modifications (i.e., without emissions controls) to attain the
control efficiency and there are no further process
modifications.
Response; Proposed subparts HH and HHH contained provisions
allowing the owner or operator to demonstrate a 95 percent HAP
emission reduction using process modifications. The EPA did not
intend to preclude owners or operators from using combinations of
process modifications and control devices. Therefore, to clarify
that owners or operators have the option of using combinations of
process modifications and control devices, §§63.765(c)(2) and
63.1275(c)(2) of the final rules are as follows:
(2) The owner or operator shall demonstrate, to
the Administrator's satisfaction, that the total HAP
emissions to the atmosphere from the glycol dehydration
unit process vent are reduced by 95.0 percent through
process modifications, or a combination of process
modifications and one or more control devices, in
accordance with the requirements specified in
§63.771(e) [for subpart HH and §63.1281(e) for subpart
HHH] .
The EPA does not agree with commenter IV-D-30's recommendation to
allow a one-time compliance demonstration. The EPA does not
believe that a one-time compliance demonstration would ensure
future or continuous compliance. Therefore, the EPA has not
included the commenter"s suggested language. Instead, the final
rules contain provisions that require owners or operators to:
(1) establish and document glycol dehydration unit baseline
operations; (2) document the conditions for which the glycol
dehydration unit baseline operations will be modified to achieve
a 95 percent overall HAP emission reduction using process
modifications or a combination of process modifications and one
or more control device; (3) maintain records demonstrating that
the facility operates under the conditions of the process
2-148
-------
modification; and (4) if a control device is used in combination
with the process modifications, demonstrate that the control
device achieves the emission reduction required for an overall
emission reduction of 95 percent [§§63.771(e) and 63.1281(e)].
Only modifications in glycol dehydration unit operations directly
related to process changes (such as glycol recirculation rate or
glycol-HAP absorbency] are allowed. Changes in gas inlet
characteristics or natural gas throughput rate are not allowed to
be used as process modifications.
Comment: Commenter IV-D-05 stated that the requirement in
§63.765(c) to reduce emissions from the reboiler vent and flash
tank by 95 percent was ambiguous because the flash tank should
not be vented. According to the commenter, all of the offgas
from the flash tank should be recovered, and that GLYCalc assumes
this. The commenter suggested that §63.765 (c) should be
clarified to state that "... HAP from a glycol process should
be reduced to 95 percent as compared with HAP without any process
modifications." The commenter noted that the term "process
modifications" would have to be defined, and the EPA would have
to address whether a flash tank is a process modification.
Commenter IV-D-05 also recommended that the definition of
GCG separator in subpart HH should state that all off-gas must be
recovered. The commenter also stated that the gas-condensate-
glycol (GCG) separator should not be called a tank since it is a
pressurized vessel.
Response; Although the GCG separator is a pressurized
vessel the industry commonly refers to it as a flash tank. For
example, the GCG separator is labeled as a flash tank in the TEG
dehydration flowsheet presented in GLYCalc. Therefore, the EPA
has not modified the definition of GCG separator in subparts HH
and HHH.
2-149
-------
The EPA does not agree with the commenter's statement that
all off-gas must be recovered and that GLYCalc assumes that all
off-gas is recovered. According to the GLYCalc Dehydration
Handbook, contained electronically within the GLYCalc program,
the flash gas from the GCG separator can be used as a
supplemental fuel gas or as stripping gas in the reboiler, but
may be vented to the atmosphere at some locations. In addition,
the emission calculation in GLYCalc plainly separates flash gas
emissions from the GCG separator from the reboiler vent
emissions.
Although the EPA has not modified the definition of GCG
separator in response to this comment, the EPA believes that the
requirements in §§63.765(c)(3) and 63.1275(c)(3) need to be
clarified. As proposed, §§63.765 (c) (3) and 63.1275 (c) (3) stated
that control of HAP emissions from the flash tank ". . .is not
required if the owner or operator demonstrates to the
Administrator's satisfaction, that total HAP emissions to the
atmosphere from the glycol dehydration unit reboiler vent and GCG
separator (flash tank) are reduced by 95 percent." These
requirements were intended to provide owners and operators the
flexibility to install a control device to control emissions from
the reboiler vent such that the emission reduction from the
glycol dehydration unit process vent (which is defined in
§§63.761 and 63.1271 to include the flash tank and the reboiler
vent) is equivalent to 95 percent. Thus, the owner or operator
would not be required to install separate control devices for the
reboiler and flash tank vents. Therefore, the EPA has modified
§§63.765(c)(3) and 63.1275(c)(3) to clarify this intent as
follows:
2-150
-------
(3) Control of HAP emissions from a 6CG separator
(flash tank) vent is not required if the owner or
operator demonstrates, to the Administrator's
satisfaction, that total HAP—emissions to the
atmosphere from the glycol dehydration unit reboiler
process vent and GCC separator—(flaoh- tonic)—veafc are
reduced by 05 percent one of the levels specified in
paragraphs (c)(3)(i) through fc)(3)(±i) of this
section, through controls as specified in paragraph
(b)(1) of this section.
fi) HAP emissions are reduced by 95.0 percent or
more.
fii) Benzene emissions are reduced to a level less
than 0.90 meoaarams per year.
Comment: Commenter IV-G-07 suggested an alternative MACT
rule based on air-cooled condensers:
1. Stream exiting an air-cooled glycol dehydrator vent
condenser shall be in vapor/liquid equilibrium at or below a
temperature of 70°F or within ten Fahrenheit degrees of the
then current air temperature, averaged over any 24-hour
period.
2. Air emissions shall be less than: 5 tons/yr of benzene, 15
tons per year of HAP, and 50 tons per year of VOC.
3. A rich glycol flash tank must be used in any glycol
dehydration system, where its use will cause that system's
vent condenser to condense and recover more than an
additional 10 tons per year of VOC.
The commenter claimed that the proposed alternative: would
eliminate safety hazards associated with forcing operators to
burn some vent streams; is technically sound and cost effective;
and is good public policy by encouraging hydrocarbon recovery.
The commenter claimed that the health risk concerns due to higher
emissions on hot days from air cooled condensers are not really
the problem, since in reality, hot daytime atmospheres are
unstable resulting in vent emissions becoming well mixed.
According to the commenter, cool, still nights are a higher
potential exposure risk, which is small in any event.
Response; Section 112 of the Act requires the EPA to
establish standards no less stringent than the MACT floor. As
2-151
-------
stated in the previous response, the EPA determined that a
control efficiency of 95 percent represented the MACT floor. The
EPA does not believe that the commenter's suggestions represent
the MACT floor.
Comment; Commenter IV-D-07 interpreted §§63.765(c) (2) and
(3) and §§63.1275 (c) (2) and (3) to mean that emissions from both
the reboiler vent and flash tank vent can be used in determining
the emission reduction. The commenter supported this option.
Response; The commenter has correctly interpreted the
requirements specified in §§63.765(c)(3) and 63.1275(c)(3).
Comment: Commenter IV-D-07 stated that the EPA should
consider the fact that still vent and flash gas streams can be
laden with water and their use as a fuel source may not be
possible. The commenter suggested that burning these streams
would not be possible for smaller or unmanned facilities since
additional natural gas may need to be added to the stream before
flaring. Additionally, according to the commenter, the stream
composition may be inconsistent, and its use as a fuel or in a
flare may need to be closely monitored.
Response; Combustion of still vent and flash gas streams is
not required by subparts HH and HHH. It is up to the owner or
operator to decide whether combustion is a viable alternative for
control. In the final subparts HH and HHH, an owner or operator
has the option of complying with: (1) a HAP emission reduction of
95 percent or more; (2) an outlet HAP concentration of 20 ppmv or
less (for combustion devices) or (3) a benzene emission limit of
1 tpy. Therefore, the owner or operator should decide which
control device to use to comply with the required reductions
depending on individual stream characteristics.
2.10.4 Equipment Leak Standards
Comment: Commenter IV-D-05 requested that §63.769(a) be
clarified if the EPA intends to include ancillary equipment at
2-152
-------
production facilities, and suggested the following wording:
"This section applies to: (1) ancillary equipment at natural gas
processing plants, and to (2) compressors (as defined in 63.761)
at natural gas processing plants that ..." The commenter also
requested that the EPA justify the fact that this program does
not afford the same leniency as subpart KKK for plants that
process less than 10 MMscf/d regarding routine monitoring.
Response; The commenter is incorrect in stating that
subpart HH does not afford the same leniency as subpart KKK for
plants that process less than 10 MMscf/d. On the contrary,
§63.769 (c) (5) exempts equipment located at nonfractionating
plants with the capacity to process 10 MMscf/d from routine
monitoring requirements, and is consistent with §60. 633 (d) of
subpart KKK. The metric capacity that is equivalent to
10 MMscf/d should be 283,000 standard cubic meters per day
(m3/day) , rather than the proposed 283 m3/day. This has been
corrected in the final rule.
With regard to the commenter ' s first request, the EPA has
made the following change to clarify applicability to §63. 769 (a):
(a) This section applies to equipment sub-iect to
located at natural as rocessin
and specified in paragraphs (a) (1) and (a) (2) of this
section, ancillary equipment and- compreooors — fee
defined in §63.761) — at natural gas processing plants
that contain or contact a fluid (liquid or gas) that
has a total VOHAPVHAP concentration equal to or greater
than 10 percent by weight (determined according to the
previsions of 40 CFR 61. 245 (d) procedures specified in
S63 .772 (a) ) and that operates in VHAP service equal to
or greater than 300 hours per calendar year.
(1) Ancillary equipment, as defined in S63.761;
and
( 2 ) Compressors .
Comment : Commenter IV-D-06 recommended that the EPA revise
§63. 769 (a) to apply only to equipment operating in VOHAP service
for 300 hours or more per year. According to the commenter, if
2-153
-------
the equipment is in operation more than 300 hours per year, but
is in VOHAP service for only a small part of the time, there
would be no need for standards to apply.
Response; The EPA agrees with the commenter, and has
modified §63.769(a) in response to this comment to clarify that
equipment operating in VHAP service for more than 300 hours per
year is subject to the rule.
Comment: Commenter IV-D-16 stated that it is confusing to
point to 40 CFR part 61, subpart V, which is unfamiliar to gas
plant operators, when subpart V is almost the same as 40 CFR part
60, subpart KKK. The commenter recommended that subpart HH point
to subpart KKK, since the regulated community and compliance
inspectors are familiar with it and will understand it better.
The commenter also noted that the MACT floor is subpart KKK.
Response; The EPA determined that the MACT floor for
equipment leaks is subpart KKK and the NSPS level of control in
subpart KKK is equal to that of 40 CFR part 61, subpart V.
However, subpart KKK is a standard that controls VOC and
subpart V controls HAP. Since the pollutants targeted for
control under subpart HH are HAP, cross-referencing the
requirements from the equipment leaks NESHAP (40 CFR part 61,
subpart V) is appropriate.
Comment: Commenters IV-D-20 and IV-D-22 recommended that it
would be less burdensome and would avoid redundancy if provisions
were added, for facilities that are subject to other federal,
State, and local LDAR programs, to allow control of equipment
leaks under similar programs. Commenter IV-D-20 urged the EPA to
allow for a process of equivalency and/or stringency
demonstration for these other requirements. The commenter also
urged the EPA to clarify that facilities have the option of
complying with only one rule that will subsume all other LDAR
requirements. Furthermore, the commenter requested that §63.769
be expanded to allow for other equivalent, or more stringent
2-154
-------
State, and local LDAR programs/rules to be used instead of
subpart HH requirements, provided the governing rule is
specifically included in the facility's title V permit.
Commenter IV-D-22 endorsed the proposed provisions intended to
prevent duplication of effort and requirements for facilities
subject to LDAR requirements in part 63, subpart H, or subpart
KKK NSPS.
Response; The EPA believes that facilities subject to other
federal, State and local regulatory programs should be allowed to
comply with the requirements those programs, if they are at least
as stringent, or equivalent to subparts HH and HHH. Sections
63.777 and 63.1287 already contain provisions for alternative
emission limitations that must be at least as equivalent as
subpart HH or HHH as appropriate.
2.10.5 Control Device Requirements
Comment: Commenters IV-D-06, IV-G-02 and IV-G-12 stated
that §§63.771(d) (5) and 63.1281 (d) (5) should only require that
spent carbon be managed as a hazardous waste if it is, in fact, a
hazardous waste. According to the commenters, it is not a listed
waste, so unless it displays a hazardous characteristic, it
should not have to be managed as a hazardous waste. Commenter
IV-D-06 recommended that the EPA should allow the option of
managing the carbon in a combustion device regulated under any
subpart of part 60, 61, or 63. The Commenter stated that this
comment may also apply to subpart HH. Further, regarding
acceptable treatment methods, commenters IV-G-02 and IV-G-12
recommended the words "for which the owner or operator" be
changed to "whose owner or operator" to make clear that it is the
treatment facility, and not the generator, who must obtain the
proper RCRA permits or interim status.
Response; The EPA agrees with the commenters and has
replaced proposed SS63.771(d)(5) and 63.1281(d)(5) with the
following:
2-155
-------
(5) For each carbon adsorption system used as a
control device to meet the requirements of paragraph
(d)(1) of this section, the owner or operator shall
manage the carbon as follows:
(i) Following the initial startup of the control
device, all carbon in the control device shall be
replaced with fresh carbon on a regular, predetermined
time interval that is no longer than the carbon service
life established for the carbon adsorption system.
(ii) All carbon removed from the control device
be managed in one of. the following manneye-t-The
shall be either regenerated, reactivated, or burned in
Qne of the units specified in paragraphs (d) (5) (ii) (A)
through (d)(5)(ii)(G) of this section.
(A) Regenerated or reactivated in a thermal
treatment unit for which the owner or operator has
either been issued a final permit under 40 CFR part
270-7—and designs and operates the unit in accordance
wartefe that implements the requirements of 40 CFR 264,
subpart X-;—or certified compliance with the interim
statue requirements of 40 CFR 265,—subpart—P.
(B) Regenerated or reactivated in a thermal
treatment unit equipped with and operating air emission
controls in accordance with this section.
(C) Regenerated or reactivated in a thermal
treatment unit equipped with and operating organic air
emission controls in accordance with a national
(D) Burned in a hazardous waste incinerator for
which the owner or operator has been issued a final
permit under 40 CFR part 270?—and designs and operates
tho unit in accordance with that implements the
requirements of 40 CFR 264, subpart O.
(E) Burned in a hazardous waste incinerator which
the owner or operator has designed and operates in
ffy v>par t 0»
4&f(F) Burned in a boiler or industrial furnace
for which the owner or operator has either been issued
a final permit under 40 CFR part 270-r-aad designs and
operates the unit in accordance with that implements
the requirements of 40 CFR part 266, subpart H.
2-156
-------
(G) Burned in a boiler or industrial furnace which
aceordanee^—or hao certified compliance with the
interim status requirements of 40 CFR part 266, subpart
H.
2.10.6 gtoragg Vessel Standards
Comment: Commenter IV-D-16 stated that external floating
roofs complying with subpart Kb need to be addressed in the
storage vessel standard, if they are allowed. Commenter IV-D-22
stated that the proposed definition of cover in §63.761 includes
an external floating roof as an example; however, storage vessel
standards in §63.766 do not list external floating roofs as an
allowed control option. In fact, the commenter noted that
§63.766(b)(1) suggests that an external floating roof would need
to be connected through a closed-vent system to a control device.
The commenter stated that they do not believe the EPA intended
this result, because other existing standards, such as the
subpart Kb New Source Performance Standards (NSPS) (40 CFR
Section SO.llOb), allow the external floating roof alone. The
commenter also stated that they do not endorse the detailed
control requirements in subpart Kb (for external or internal
floating roofs) for exploration and production (E&P) storage
tanks. In particular, the provisions in subpart Kb for the many
vents, fittings, lids, and other equipment on both internal and
external floating roofs are inappropriate for oil exploration and
production tanks. According to the commenter, it is not
appropriate to implement controls on E&P tanks that are more
stringent than the controls for tanks in a refinery and it is not
supported by the MACT floor for production tanks. Additionally,
the commenter noted that subpart Kb does not apply to the small
vessels typically located at production facilities.
Response; The EPA did not intend to limit the types of
covers allowed to only those listed as examples in the definition
of cover in §63.761. Therefore, the EPA has added language to
the definition of cover in §63.761 as follows: "...Examples of a
2-157
-------
cover include, but are not limited to. a fixed-roof installed on
a tank, an external floating roof installed on a tank, and a lid
installed on a drum or other container."
The EPA1s data, collected from section 114 questionnaires
and site visits indicated that the control technology in use to
control existing storage vessels did not include internal or
external floating roofs. Therefore, the EPA. has removed the
requirements for internal floating roofs contained in proposed
§63.766(b)(3). However, in order to allow owners or operators
the option of complying with the requirements specified in 40 CFR
part 60, subpart Kb, 40 CFR part 63, subpart G, or 40 CFR part
63, subpart CC the EPA has added the following paragraph to
§63.766:
(d) This section does not apply to storage vessels
for which the owner or operator is meeting the
requirements specified in 40 CFR part 60, subpart Kb;
or is meeting the requirements specified in 40 CFR part
63, subparts G or CC.
2-158
-------
2.11 MONITORING, RECORDKEEPING, AND REPORTING
Comment: Commenters IV-D-08, IV-D-22, and IV-D-34 requested
that the EPA allow oversized control devices to be exempt from
monitoring, inspection, recordkeeping and reporting requirements
other than the initial design analysis. The commenters stated
that oversized devices will essentially always meet the
regulatory requirements. According to the commenters, if a
design analysis shows that the device is oversized so that
compliance with the control efficiency requirement will be met
even during worst case conditions, the exemption should be
allowed. The commenters explained that an exemption would allow
the owner or operator to spend more on the device and less on
monitoring, inspection, recordkeeping, and reporting over the
life of the facility. The commenters further stated that is it
not cost-effective to continue monitoring, inspection,
recordkeeping, and reporting for a device that will always meet
regulatory requirements.
Response; Monitoring, inspection, recordkeeping and
reporting requirements ensure continuous compliance with the
standards. Over-designing a control device would not ensure
proper operation and compliance with the standards. Therefore,
the EPA has not modified subparts HH and HHH in response to these
comments.
Comment: Commenters IV-D-20 and IV-D-22 urged the EPA to
reevaluate how the monitoring, recordkeeping, and reporting
requirements can be made clear, nonoverlapping, and implementable
in the field. According to the commenters, the "cut and paste"
approach apparently used by the EPA to develop the monitoring,
recordkeeping, and reporting requirements leads to a burdensome
set of confusing, and sometimes unnecessary, requirements.
Commenter IV-D-22 stated that they believe the proposed
monitoring, recordkeeping, and reporting requirements impose a
significant burden on E&P facilities that cannot be justified
2-159
-------
based on any current monitoring, recordkeeping, and reporting for
oil or gas facilities.
Commenter IV-D-09 noted that §§63.10(b)(2) and 63.10(c)
require 24 separate record entries for each dehydration unit and
control device, as well as up to seven separate reports. The
commenter stated that many of these reports are superfluous. As
an example, the commenter suggested consolidating all information
required to ensure the operational status of the monitoring
system into one combined maintenance and operational log. The
commenter also requested that the EPA work with operators to
explore ways to incorporate the information required for
compliance demonstrations into existing recordkeeping and
reporting practices realistically.
Response; The EPA recognizes that unnecessary monitoring,
recordkeeping, and reporting requirements would burden both the
source and enforcement agencies. Prior to proposal, the EPA
attempted to reduce the amount of monitoring, recordkeeping, and
reporting to only that which is necessary to demonstrate
compliance.
In response to the commenters' concerns, the EPA reevaluated
whether monitoring, recordkeeping, and reporting requirements
could be further reduced while maintaining the enforceability of
the rule. Therefore, the EPA has made the following changes in
the final rules (subparts HH and HHH) to further reduce the
monitoring, recordkeeping, and reporting burden.
(1) Almost all reports have been consolidated into the
Notification of Compliance Status Report and the Periodic
Reports.
(2) If multiple tests are conducted for the same kind of
emission point, using the same test method, only one complete
test report is submitted along with the summaries of the results
of other tests.
(3) Site-specific test plans describing quality assurance
in §63.7(c) of 40 CFR part 63, subpart A are not specifically
required in the individual subparts because the test methods
2-160
-------
cited in subparts HH and HHH already contain applicable quality
assurance protocols. It should be noted that the Administrator
would still have the authority to request a test plan.
(4) Periodic reports are required to be submitted
semiannually for all facilities (the proposal required quarterly
reports if monitored parameters were out of range more than a
specified percentage of time).
(4) A reduction in the record retention requirements for
monitored parameters. The proposal required values of monitored
parameters to be recorded every 15 minutes and all 15-minute
records had to be retained. The final rule requires monitored
parameters to be recorded every hour and all hourly records to be
retained.
Comment: Commenter IV-D-38 suggested that §63.764 (c) be
modified as follows so that affected sources that are not at
major HAP sources do not have to meet stringent control,
monitoring, and recordkeeping requirements:
(c) Except as specified in paragraph (e) of this
section, the owner or operator of an affected source
located at an existing or new major HAP source shall
comply with the standards in this subpart as specified
in paragraphs (c)(1) through (c)(3) of this section.
Response; A major source is defined in §63.2 as "any
stationary source" that "emits, or has the potential to emit,
. . , any hazardous air pollutant." Although the EPA believes
that specifying that affected sources are located at existing or
new major HAP sources would be redundant, the EPA has made the
suggested modification.
2.11.1 Monitoring Requirements
Comment: Commenter IV-D-01 questioned the basis for the
10,000 ppm leak definition in §63.769 (c) (2) and requested that
the 10,000 ppm leak definition for pressure relief devices in
gas/vapor service be changed to 500 ppm. The commenter stated
that the pressure relief devices in gas/vapor service should have
a leak definition of 500 ppm above background in accordance with
40 CFR 61.242-4(a).
Commenter IV-D-01 also requested that the EPA delete the
provision for which pressure relief devices, in a
2-161
-------
nonfractionating facility monitored only by non-facility
personnel, may be monitored after a pressure release the next
time the monitoring personnel are on-site, instead of within five
days (not to exceed 30 days following a pressure release without
monitoring). The commenter stated that the owner or operator
could make provisions for either company- personnel or contract
personnel to perform the monitoring. The commenter also stated
that pressure release devices should be monitored no later than
five calendar days after each pressure release in accordance with
40 CFR 61.242-4(b)(2).
Response; Currently, the oil and natural gas production
industry is regulated by the NSPS for equipment leaks of VOC from
onshore natural gas processing plants (40 CFR part 60, subpart
KKK) . The EPA determined that the MACT floor for equipment leaks
at natural gas processing plants was the NSPS level of control.
Subpart KKK requires owners and operators to monitor pressure
relief devices quarterly and within five days after each pressure
release to detect leaks [§63.633(b)(3)], with a leak defined as
an instrument reading of 10,000 ppm or greater [§60.633(b) (2)] .
Section 61.242-4(a) requires pressure relief devices to be
operated with no detectable emissions, which is more stringent
than subpart KKK. Since the MACT floor was determined to be
equivalent to the level of control specified in subpart HHH, the
EPA has not changed the leak definition for pressure relief
devices, as requested by the commenter.
The requirement allowing the owner and operator of a
nonfractionating facility, which are monitored only by
non-facility personnel, to monitor after a pressure release the
next time the monitoring personnel are on-site is consistent with
40 CFR part 60, subpart KKK [§63.633(b) (4)]. Since the MACT
floor was determined to be the level of control required by
2-162
-------
subpart KICK, the EPA has not made the change suggested by the
connnenter.
Comment; Commenter IV-D-22 recommended that the EPA allow
for a deviation from a 2 percent leak rate during startup,
shutdown, or malfunction, per the facility startup, shutdown and
malfunction plan, as specified in §63.10(d)(5), without forgoing
the option for skip monitoring. Although subpart HH does not
mention skip monitoring directly, the commenter stated that they
believe it is invoked through the reference to the leak detection
and repair (LDAR) work practice standard in 40 CFR part 60,
subpart W. [Note: The commenter's citation is incorrect.
Subpart HH refers to 40 CFR part 61, subpart V.]
Response; The purpose of the startup, shutdown, and
malfunction (SSM) plan is to ensure that owners and operators
operate and maintain affected sources with good air pollution
control practices at all times. The SSM plan also ensures that
owners or operators are prepared to correct malfunctions quickly,
to minimize excess HAP emissions. The EPA believes that
monitoring is warranted during SSM, however, the EPA has modified
§63.774(b)(3) to state that the leaks that occur during SSM
events do not count toward the percent leak rate, provided the
SSM plan is followed.
Comment: Several commenters were concerned with the
provisions specifying the accuracy of the measurement devices
used to comply with the subpart. Commenter IV-D-06 recommended
that the EPA allow measurement devices with better accuracy than
what subpart HHH requires. The commenter contended that the way
subpart HHH was written, using devices with better accuracy than
the rule specifies is forbidden. The commenter suggested that
the EPA revise the following sections and paragraphs to allow
more accurate measuring devices. The commenter noted that there
may also be other paragraphs (in particular, they did not look
2-163
-------
closely at the compliance demonstration requirements) and if so,
the EPA should revise them similarly to the examples shown below.
Section 63.1271:
Temperature monitoring device means a unit of
equipment used to monitor temperature and having art
minimum accuracy of ± I percent of the temperature
being monitored expressed in °C, or ± 0.5°C, whichever
is greater.
Section 63.1282(a)(1)(i):
(i) The owner or operator shall install and
operate a monitoring instrument that directly measures
flow to the glycol dehydration unit with an accuracy of
plus or minus 2 percent or better.
Section 63.1283 (d) (3)'(i) (A), (B) , (D) , (E) , and (F) :
(A) For a thermal vapor incinerator, a temperature
monitoring device equipped with a continuous recorder.
The monitoring device shall have an minimum accuracy of
±1 percent of the temperature being monitored in °C ,or
±0.5 °C, whichever value is greater. The temperature
sensor snail be installed at a location in the
combustion chamber downstream of the combustion zone.
(B) For a catalytic vapor incinerator, a
temperature monitoring device equipped with a
continuous recorder. The device shall be capable of
monitoring temperature at two locations and have an
minimum accuracy of ±1 percent of the temperature being
monitored in °C, or ±0.5 °C, whichever value is
greater. One temperature sensor shall be installed in
the vent stream at the nearest feasible point to the
catalyst bed inlet and a second temperature sensor
shall be installed in the vent stream at the nearest
feasible point to the catalyst bed outlet.
(D) For a boiler or process heater with a design
heat input capacity of less than 44 megawatts, a
temperature monitoring device equipped with a
continuous recorder. The temperature monitoring device
shall have an minimum accuracy of ±1 percent of the
temperature being monitored in °C, or ±0.5 °C,
whichever value is greater. The temperature sensor
shall be installed at a location in the combustion
chamber downstream of the combustion zone.
(E) For a condenser, a temperature monitoring
device equipped with a continuous recorder. The
temperature monitoring device shall have an minimum
accuracy of ±1 percent of the temperature being
monitored in °C, or ±0.5 °C, whichever value is
2-164
-------
greater. The temperature sensor shall be installed at
a location in the exhaust vent stream from the
condenser.
(F) For a regenerative-type carbon adsorption
system, an integrating regeneration stream flow
monitoring device equipped with a continuous recorder,
and a carbon bed temperature monitoring device equipped
with a continuous recorder. The integrating
regeneration stream flow monitoring device shall have
an minimum accuracy of ±10 percent and measure the
total regeneration stream mass flow during the carbon
bed regeneration cycle. The temperature monitoring
device shall have an minimum accuracy of ±1 percent of
the temperature being monitored in °C, or ±0.5 °C,
whichever value is greater and measure the carbon bed
temperature both after the regeneration and within 15
minutes of completing the cooling cycle, and over the
duration of the carbon bed steaming cycle.
The commenter stated that this comment also applies to subpart
HH.
Response; The EPA agrees with the commenter's
recommendation to modify the monitoring device accuracy
specifications to state that the accuracy requirements are the
minimum necessary to demonstrate compliance and has modified the
subparts HH and HHH as suggested.
Comment: Commenters IV-D-08 and IV-D-22 requested that the
EPA require flow instrumentation with an accuracy of ±2 percent
only when the measured values are within 98 percent of the
exemption or compliance targets. The commenters stated that
while this level of accuracy is available, proving compliance for
streams that do not have flows close to the exemption or
compliance levels is not necessary. Commenter IV-D-22 stated
that throughout §63.773, subpart HH refers to temperature
monitoring devices with an accuracy of ± 1 percent of the
temperature being monitored in °C, or ±0.5 °C, whichever value is
greater. The commenter, along with commenter IV-D-08, stated
that they do not believe that the MACT floor for this source
category supports the accuracy of 2 percent for measuring flow
and I percent for measuring temperature, or that they have
demonstrated continuous compliance with applicable standards.
2-165
-------
Commenter IV-G-12 also stated that the ±0.5 °C measurement is
practically useless for many analog temperature recorders.
Commenter IV-D-22 stated that requiring a unit to demonstrate
compliance with a temperature requirement is unnecessary and
unduly restrictive unless the operator chooses to select a
temperature set point within 1 percent of the requirement. The
commenter recommended, and commenter IV-D-34 supported, that
subpart HH be modified to eliminate specified accuracies for flow
and temperature measurement devices and to allow whatever
accuracy is necessary to demonstrate compliance with the
temperature or flow target.
Response: The average emission limitation achieved by the
top 12 percent of the facilities (for source categories with more
than 30 facilities) has to be considered for the MACT floor for
existing sources. The EPA believes that accuracy requirements
are necessary to demonstrate ongoing compliance. The EPA also
believes setting site-specific accuracy requirements would be
unduly burdensome for the permitting agencies. In addition, a
minimum accuracy provides an advantage for the owner or operator
because they would not be required to use the same monitor
forever. Furthermore, if the accuracy requirements were removed,
additional recordkeeping and reporting requirements would be
necessary to ensure that less accurate monitors were not
installed after the performance tests. However, to provide
additional flexibility in selecting monitoring devices, the EPA
has changed the accuracy levels from ± 1 percent of the
temperature being monitored, in °C or ± 0.5 °C, to ± 2 percent of
the temperature being monitored, in °C or + 2.5 °C, whichever is
greater.
Comment: Commenters IV-D-22 and IV-D-34 recommended that a
provision be added to allow design analysis or engineering
2-166
-------
calculations in place of monitoring if they can demonstrate that
limits will not be exceeded.
Response; Allowing design or engineering calculations would
show theoretical compliance, but would not demonstrate continuous
compliance with the emission standard. The EPA believes that the
monitoring requirements in subparts HH and HHH are the minimum
necessary to ensure compliance with the standards. Therefore,
the EPA has not added the provisions recommended by the
commenters.
Comment: Commenter IV-D-06 recommended that the EPA clarify
the bypass monitoring requirements. The commenter stated that
§63.1281(c) (3) (i) (A) says flow indicators must indicate "whether
gas, vapor, or fume flow is present" at least once every 15
minutes. The commenter suggested that the wording seems to
require a direct indication of "flow." The commenter requested
that the definition of "flow indicator" be revised to allow valve
position indicators. The commenter stated that valve position
indicators will not give a reading of whether flow is present,
but they will give a reading of whether a diversion has occurred.
In addition, the commenter noted that §63.1281 (c) (3) (i) (B) says
that car-seals are intended to show that valves are in the
"closed" position. The commenter provided an example of a
three-way valve that is commonly used in bypass situations.
According to the commenter, three-way valves have two "open"
positions and one "closed" position. In those cases, the valve
should be car-sealed in the open position that goes to the
control device. The commenter felt that the literal wording of
subpart HHH does not allow that option. The commenter
interpreted the wording to read that the "closed" position is
required. The commenter suggested the following changes to
§63.1281(c) (3) (i) (A) and (B) .
(A) Install, calibrate, maintain, and operate a
flow indicator at the inlet to the bypass device that
indicates takes a reading at least once every 15
2-167
-------
minutes whethergas,vapor,
tlig bypass device; or
(B) Secure the valve installed at the inlet to the
bypass device in the closed non-diverting position
using a car-seal or a lock-and-key type configuration.
The owner or operator shall visually inspect the seal
or closure mechanism at least once every month to
verify that the valve is maintained in the closed
non-diverting position.
The commenter stated that this comment also applies to subpart
HH.
Response; The EPA believes that clarifications to the
bypass requirements are necessary to allow valve position
indicators, and to require that a car-seal be used to secure a
valve installed at the inlet to the bypass device in the
non-diverting position. The EPA has modified §§63.771(c) (3) (i)
and 63.1281(c)(3)(i) as follows:
(A) Properly install, calibrate, maintain, and
operate a flow indicator at the inlet to the bypass
control device to the atmosphere that indicates takes a
reading at least once every 15 minutes and sounds an
alarm when the bypass device is open such that the
control device to the atmospherewhether gas.—vapor,—ea?
fume flow io present in the bypaoo device; or
(B) Secure the bypass device valve installed at
the inlet to the bypass device in the closed
non-diverting position using a car-seal or a
lock-and-key type configuration. The owner or operator
shall visually inspect the seal or closure mechanism at
least once every month to verify that the valve is
maintained in the e4«eed non-diverting position and the
vent styj'?lTn is not diverted through the bypass device.
In addition, the final rules contain a definition of flow
indicator in §§63.761 and 63.1271.
The EPA has also added recordkeeping and reporting
requirements for the bypass line requirements. The final rules
2-168
-------
contain recordkeeping requirements [§§63.774(b)(4)(iii) and (iv)
and 63.1284(b)(4)(iii) and (iv)], which require an owner or
operator to maintain records of whether the flow indicator was
operating and whether flow was detected at any time during the
hour, as well as records of the times and durations of all
periods when the vent stream is diverted from the control device
or the monitor is not operating. When a seal or closure
mechanism is used, hourly records of flow are not required. In
such cases, the owner or operator is required to record that the
monthly visual inspection of the seals or closure mechanism has
been done, and shall record the duration of all periods when the
seal mechanism is broken, the bypass line valve position has
changed, or the key for a lock-and-key type lock has been checked
out, and records of any car-seal that has broken.
The final rule also requires owners or operators to include,
in the periodic reports, all periods when the vent stream is
diverted from the control device through a bypass line. When a
seal or closure mechanism is used, the periodic report must
contain all periods in which the seal mechanism was broken, the
bypass valve position was changed, or the key to unlock the
bypass line valve was checked out.
In addition, periods where the vent stream has been diverted
through the bypass line, as indicated by the flow indicator or
the closure mechanism or seal, are defined as excursions except
when they occur during startup, shutdown, or malfunction events
or periods of nonoperation.
Comment: Commenter IV-D-06 requested that the EPA revise
Table 2 to say that §63.8(e) does not apply to subpart HHH. The
commenter stated that subpart HHH does not require a performance
evaluation at a specific time on a continuous monitoring system.
2-169
-------
Therefore, according to the commenter, if §63.8(e) applies,
nothing is required. The commenter further stated that various
other MACT standards have not required performance evaluations on
continuous monitoring systems. The commenter suggested that
imposing the burden of these performance evaluations is
unnecessary. The commenter stated that if subpart HH also
incorporates §63.8(e), then this comment also applies to subpart
HH.
Response; Although performance evaluations on continuous
monitoring systems are not specifically required under subparts
HH and HHH, the Administrator retains the authority to request
such evaluations. Therefore, the EPA has modified table 2 of
subparts HH and HHH to state that the applicable subpart does not
specifically require continuous monitoring system performance
evaluations but that the Administrator can request the owner or
operator to conduct performance evaluations.
Comment: Commenter IV-D-31 requested that if the intent of
the monitoring protocol is to ensure that the equipment is
operating correctly, then subpart HHH should state that intent.
The commenter also requested that the EPA consider the complexity
of the control equipment when designing a monitoring period. The
commenter suggested that the frequency of monitoring be based on
the probability of a device failing to function. The commenter
suggested that simple condenser systems cannot vary widely in
performance over short periods and have a minimum probability of
failure, meaning that recording information at frequent intervals
is inefficient and wasteful.
Response; The intent of the monitoring protocol is to
ensure compliance with the standards. Based on the comments
received on the proposed rules, the EPA reevaluated the
monitoring requirements contained in subparts HH and HHH. As
stated in section 2.10.2 of this document, the final rule
specifies that control devices must determine compliance using
2-170
-------
the daily average of the monitored parameters. Owners and
operators that install condensers have the option of using a
365-day average tinder final subpart HH and a 30-day average under
final subpart HHH. To reduce the number of data points required
to be used in these average calculations, the EPA has modified
the monitoring requirements for control devices. The final rule
requires continuous monitoring systems to measure data values at
least once every hour and record either each measured data value,
or each block average for each 1-hour period or shorter periods
calculated from all measured data values during each period.
Comment: Commenter IV-D-06 stated that components exempt
from instrumental leak detection monitoring in §63.1283 (c) (3)
(such as components under a vacuum) should also be exempt from
the initial monitoring. The commenter stated that this comment
may also apply to subpart HH. Commenter IV-D-22 suggested that
the type of "no detectable emissions" monitoring required in
§§63.771(c) (2) and 63.773 (c) (1) (ii) cannot be justified by a MACT
floor analysis.
Response; The EPA believes that initial monitoring is
necessary to show that the closed-vent system is has been
designed to operate with no detectable emissions. However, the
EPA believes that once closed-vent system components that are
permanently or semi-permanently sealed (e.g., welded joints) have
been shown to operate with no detectable emissions, future
monitoring is not necessary, unless components are repaired,
replaced or unsealed. Therefore, the final rule requires annual
visual inspections for these components to detect defects that
could result in air emissions. However, the final rule requires
closed-vent system components that are not permanently or
semi-permanently sealed to be monitored annually to demonstrate
2-171
-------
that they are operated with no detectable emissions, in addition
to the initial inspection and the annual visual inspections.
Comment: Commenter IV-D-07 supported the EPA's statement
that "[t]he CMS that uses gas chromatography to measure
individual organic HAP compound chemicals is not practical for
applications where multiple organic HAP chemicals are to be
monitored ..." According to the commenter, monitoring of
control devices should be flexible to allow for impracticalities
in using CMS. Furthermore, the commenter supported the
monitoring of control device operating parameters' performance
for compliance demonstrations.
Response; The EPA appreciates the commenter's support. As
stated in the preamble to the proposed rules (63 FR 6307), the
EPA rejected the use of continuous emission monitoring systems
(CEMS) for two reasons: (1) GEMS that use gas chromatography to
measure individual gaseous organic HAP are not practical for use
when multiple HAP are monitored, and (2) CEMS that measure total
VOC or total hydrocarbons do not provide a quantified level of
the organic species present. Therefore, the EPA selected
parameters that would indicate air emission control performance
for the monitoring approach. The EPA believes that the selected
parameters are good indicators of control device performance and
continuous parameter monitoring instrumentation is available at a
reasonable cost.
Comment: Several commenters were concerned with the impact
of the monitoring requirements on remote and/or unmanned
facilities. The commenters suggested that lack of electricity,
instrument air, and personnel availability would cause problems
complying with the monitoring requirements. Commenter IV-D-08
stated that obtaining electricity for instrumentation necessary
to implement continuous monitors is expensive and burdensome for
remote facilities. Commenter IV-D-11 stated that their
facilities are located in the wetlands or in state waters and
2-172
-------
pose unique problems. The commenter explained that many of these
locations are unmanned or are manned for a few hours per day and
access to most of these facilities is by boat and sometimes by
helicopter. The commenter also noted that the weather environment
is not conducive to maintaining sophisticated electronic
equipment that affects the cost of any control strategy imposed.
According to the commenter, temperatures range from freezing in
the winter to the high 90's in the summer, with high humidity and
facilities near the coast experience corrosion problems from the
saltwater.
Response; Although the EPA has not removed the monitoring
requirements for unmanned or remote facilities, the EPA did
evaluate the possibility of reducing the requirements for
unmanned facilities.
Several of the facilities visited during the proposal during
the development of the proposal were remote and sometimes
unmanned facilities that operate through the use of automatic
control in monitoring systems. In particular, one site did not
have electrical lines to the site (II-B-2). Power was provided
by solar panels and associated storage batteries.
The EPA believes that monitoring devices are essential at
unmanned sites to ensure that control devices are operating to
ensure compliance and are the minimum necessary to ensure that
control devices are operating to ensure compliance. Therefore,
the EPA has not reduced the monitoring device requirements for
unmanned facilities.
Comment: Commenter IV-D-16 noted that most available,
reliable, accurate monitoring equipment requires electrical
power. The commenter recommended that §63.773(d) be amended to
allow for temperature indicators that do not record data. The
commenter suggested a compliance determination process: upon
inspection of the facility by a regulatory official, the
2-173
-------
temperature indicator would be observed and the reading compared
with the value chosen for the control device. If this value is
unsatisfactory, the facility operator would be required to
perform a compliance test per §63.772(e). Compliance with the
applicable standard would be based on the result of the
compliance test.
Response; The purpose of a temperature indicator is to
ensure compliance with the standard. As stated in a previous
response, the EPA considered requiring continuous monitoring
systems that measure emissions. However, the EPA determined that
parametric monitoring systems would be less burdensome but would
be good indicators of control device performance. Furthermore,
monitoring device records provide inspectors a means for
determining whether or not a control device was operating in
compliance. The EPA believes that temperature indicators that do
not record data would not give any indication of compliance over
the appropriate averaging period. Therefore, the EPA has not
modified the monitoring device requirements in response to this
comment.
Comment: Commenters IV-D-20, IV-D-22, and IV-D-34 requested
that the EPA amend §63.771(b) and (c) to require periodic visual
inspection rather than "no detectable emissions" for covers and
closed vent systems. Commenters IV-D-20 and IV-D-22 suggested
that repairs would be made if there was visual evidence of a
defect that could result in emissions. The commenters explained
that many E&P field operations have limited manpower and are
remote which prevents them from being attended daily. The
commenters stated that a "no detectable emissions" requirement
was inappropriate for exploration and production storage vessels
and requested that the requirement be dropped. According to the
commenters, the requirements would impose significant burdens on
the industry. Commenter IV-D-22 suggested that the type of "no
2-174
-------
detectable emissions" monitoring required in §§63.771(b) and
63.773(b)(3)(ii) cannot be justified by a MACT floor analysis.
In addition, commenters IV-D-20, IV-D-22 and IV-D-34 stated
that removing material from a leaking tank until the leak is
repaired [as required in §63.771(b) (2)] would result in higher
emissions during the transfer operations than would be emitted if
the material was left in place. The commenters were concerned
that removing the material from the tank would require several
wells to be shut in and would risk reservoir damage. Commenters
IV-D-20 and IV-D-22 indicated that this requirement is
inconsistent with §63.773(b)(3)(vii), which allows the material
to remain in the tank if the leak cannot be repaired within 15
calendar days after the leak is detected. The commenters urged
the EPA to delete these provisions.
Response: The EPA has evaluated the requirements for
storage vessels in subpart HH and determined that the proposed
requirements were not appropriate for the oil and natural gas
industry. Therefore, the EPA has modified §63.771(b) as follows:
(b) Cover requirements.
(1) The cover and all openings on the cover (e.g.,
access hatches, sampling ports, and gauge wells) shall
be designed to form a continuous barrier over the
entire surface area of the liquid in the tankoperato
with no detectable emissions when all cover openings
are secured in a closed,—scaled position.
cover eperateo with no detectable emissions by teoting
each opening on the cover in accordance with the
procedures specified in §63.772(e)—the first time
material is placed into the unite on which the cover is
inotallod. If a leak is detected and cannot be
repaired at the feiae that the leak is detected/—fcbe
material shall be removed from the unit and the unit
shall not be used until the leak IB repaired.
(3-2) Each cover opening shall be secured in a
closed, . . .
Additionally, the inspection and monitoring requirements
contained in proposed §63.773(b) have been combined with the
2-175
-------
inspection and monitoring requirements for closed vent systems in
§63.773(c). Therefore, §63.773(c) of final subpart HH and
§63.1282(c) of final subpart HHH (Cover and closed-vent systems
inspection and monitoring requirements) specifies that covers
must be visually inspected following the installation of the
cover, and annually for defects that could result in air
emissions.
In response to the commentera concern about the removal of
material from a leaking storage vessel, the EPA has revised the
requirements for the repair of leaks. The EPA believes that a
15-day period is sufficient to repair a leak. However, the EPA
has provided an option for a delay of repair if a repair is
technically infeasible without a shutdown, or if emissions from
the immediate repair would be greater than the fugitive emissions
resulting from the delay of repair. Therefore, §§63.773 (c) and
63.1283(c) of the final rules specify that leaks from covers or
closed-vent systems that are detected during the periodic
inspections must be repaired no later than 15 days after the leak
is detected, unless a delay of repair is requested. In this
case, the repair must be completed by the end of the next
shutdown.
Comment: Commenter IV-G-02 asked why §§63.771(d) (3) (i) (C)
and 63.1281 (d) (3) (i) (C) exempt a vent stream introduced with the
primary fuel from performance test requirements. The commenter
questioned whether the EPA has data showing that, in oil and gas
production and natural gas transmission and storage facilities,
any boiler or process heater achieves a specific HAP reduction
efficiency, or that vent streams introduced with the primary fuel
are more likely to achieve reduction and do not need a
performance test. The commenter maintained that if the EPA does
not have HAP reduction data for boilers and process heaters in
2-176
-------
this source category, the control should be a work practice for
which a performance test is unnecessary. The commenter cited
§63.644 of subpart CC (the petroleum refinery MACT) of 40 CFR
part 63 as a precedence for making the control for such vent
streams a work practice for which the monitoring requirement is a
certification that the stream is introduced into the flame zone
of the boiler or process heater.
Commenter IV-D-22 stated that §§63.771(d)(3)(i)(B) and
63.773(d)(2)(ii) exempt boilers or process heaters from
performance testing and monitoring if they have an input capacity
equal to or greater than 44 megawatts (MW). The commenter was
not aware of any fuel gas combustion devices in this source
category as large as 44 MW [(150 million British thermal units
per hour, (MMBtu/hr)]. The commenter stated that they appreciate
the EPA's attempt to reduce the compliance requirements for
certain situations, but this heat input threshold designation is
virtually meaningless for this source category.
Commenter IV-D-35 referred to §63.771(d) (1) (i) (C) , which
requires the vent stream to be introduced into the flame zone of
a boiler or process heater, and §63.771(d)(3)(i)(C), which
exempts boilers or process heaters from performance testing if
the vent stream is introduced with the primary fuel. The
commenter suggested that the wording in §63.771(d) (1) (i) (C) be
modified to require the vent stream to be mixed with the primary
fuel prior to introduction to the burner nozzles or injectors.
[Note: The reference in this sentence was §63.771(d)(3)(i)(C)
but the context indicates that it should be §63.771(d) (1) (i) (C) .
The commenter stated that boilers and process heaters that
introduce the vent stream through their own nozzles or injectors
should be subject to performance testing because the degree of
combustion being provided by the vent stream would be unknown
without performance testing.
Response; The EPA believes that the exemption for boilers
with an input capacity greater than 44 MW is important even if
2-177
-------
only one boiler in the entire source category meets the
exemption. The EPA has not removed this exemption.
The EPA's information shows that boilers or process heaters
larger than 44 MW (150 MMBtu/hr) typically operate at
temperatures and residence times necessary to achieve 95-percent
reduction or greater (usually greater than 98 percent)/ while
boilers and process heaters smaller than 44 MW are frequently not
operated to achieve the 95-percent requirement. In addition,
analyses also show that when the vent streams are introduced into
the flame zone, over 95-percent reduction is achieved. This is
because the required residence time is decreased because of the
relatively high temperature and turbulence of the flame zone.23
Additionally, the final rules do not require an initial
performance test or monitoring for boilers or process heaters
with minimum heat inputs of 44 MW, or boilers and process heaters
smaller than 44 MW if the vent stream is introduced with the
primary fuel.
Comment: Commenter IV-G-09 disagreed with the requirement
in proposed §63.1282(d) that vent streams from dehydration
condensers must be measured for HAP even if the vent streams are
introduced with combustion air as secondary fuel. The commenter
suggested that the operator is being penalized for the
environmentally appropriate step of using these streams as
secondary fuel and sampling data is meaningless. The commenter
requested that this requirement be removed. [Note: this comment
also applies to proposed §63.772(e).]
23 Hazardous Air Pollutant Emissions from Process Units in
the Synthetic Organic Chemical Manufacturing Industry -
Background Information for Proposed Standards. Volume IB:
Control Technologies. November 1992. EPA-453/D-92-016b. pg.
2-18.
2-178
-------
Response; The EPA does not agree with the commenter. As
stated in the previous response, the EPA's information indicates
that for combustion units where the vent stream is introduced
into the flame zone with the primary fuel achieve at least a
95-percent emission reduction. Although it is possible for an
individual enclosed combustion device to achieve a 95-percent
emission reduction when the vent stream is introduced with the
combustion air as secondary fuel, the EPA does not have the test
data to support this and therefore, it is not appropriate to
ensure compliance. A design analysis is allowed as an
alternative to demonstrate compliance if the owner or operator
determines that a performance test is too burdensome. Therefore,
the EPA has not modified subparts HH and HHH in response to this
comment.
Comment: Commenter IV-D-22 stated that since proposed
§63.771(d)(3)(i)(A) exempts flares from performance testing,
combustion devices such as heater treaters and glycol reboiler
burners should also be exempt from performance testing.
Response; The final rules provide an option for owners and
operators to demonstrate compliance with the 95-percent HAP
emission reduction using a flare designed and operated according
to §63.11(b). Therefore, the final rule clarifies that only
flares designed and operated in accordance with §63.11(b) are
exempt from the performance test requirements. As stated in the
previous response, boilers or process heaters smaller than 44 MW
(e.g., heater treaters or glycol reboiler burners) are frequently
not operated to achieve a 95-percent emission reduction.
Therefore, the EPA has not exempted heater treaters and glycol
reboilers from the performance testing requirements.
Comment: Commenter IV-D-22 noted that §63.773(d)(2) exempts
control devices from monitoring in which vent streams are
2-179
-------
introduced with the primary fuel. The commenter suggested that
§63.773(d) (2) (i) should be clarified to provide that "introduced
with the primary fuel" means "at the same location in the process
heater," such as the burner block, and does not require the vent
stream to be compressed and introduced into the primary fuel gas
line. No combustion or destruction efficiency advantage exists
from mixing the vent gas into the primary fuel line versus the
burner block because the vent gas undergoes the same residence
time and temperature history in either case. However,
significant additional cost is incurred to compress the vent gas
to introduce it in the primary fuel line instead of at the burner
block.
Response; The performance test and monitoring requirement
exemptions for boilers and process heaters for which the vent
stream is introduced with the primary fuel or as the primary
fuel, are consistent with the EPA's approach for several other
rules (e.g., the HON). The commenter did not provide technical
information to demonstrate that their recommendation would
provide an equivalent level of control warranting an exemption.
Therefore, the EPA has not modified subparts HH and HHH in
response to this comment.
Comment: Commenter IV-G-02 noted that in the proposed
§63.769 standards for equipment leaks, there is no exemption in
paragraphs (c)(1) through (3) from monitoring for pressure relief
devices routed through a closed vent system to a control device
(such as a flare). According to the commenter, pressure relief
devices controlled in this way cannot be monitored as the section
requires. The commenter pointed to the equipment leaks NSPS
(40 CFR part 60, subpart W), to which all NSPS for equipment
leaks (including subpart KKK) reference, which recognizes this
and provides an exemption [the commenter cited §60.482-4 (c) and
§61.242-4(c)]. Since this proposal recognizes that equipment
complying with subpart KKK of part 60 must meet more stringent
standards and are exempt from meeting §63.769 requirements, the
2-180
-------
commenter requested a similar exemption be included for units not
required to comply with the subpart KKK standards.
Response; The commenter's statement that §63.769 does not
contain monitoring exemptions for pressure relief devices routed
through a closed-vent system to a control device is incorrect.
According to §63.769 (c), an owner or operator of ancillary
equipment and compressors subject to the equipment leak standards
of subpart HH must comply with the requirements of 40 CFR 61.241
through 247. Thus, the exemption contained in §61.242-4, for
pressure relief devices routed through a closed-vent system to a
control device, applies. Therefore, the EPA did not modify
subpart HH in response to this comment.
Comment: Commenter IV-D-14 asked whether the EPA would
place additional emphasis on compliance monitoring for "synthetic
minor sources" or minor sources near the major source threshold,
as has been seen during the implementation of the title V
program.
Response; The EPA is uncertain what the commenter means by
"additional emphasis on compliance monitoring for synthetic
minors," but notes that the final rules do not alter the part 70
requirements for monitoring of any applicable requirement at
title V sources.
2.11.2 Recordkeeping and Reporting Requirements
Comment: Two commenters were concerned with the relevance
of the records required by the standard. Commenter IV-D-05 noted
that glycol units that are exempt from the control standards are
only required to keep records of gas flow rates. Based on their
experience with the results of GLYCalc 3.0, the commenter stated
that glycol circulation rates and gas composition are more
critical parameters affecting emissions than gas flow rates.
Commenter IV-D-06 suggested that keeping records of glycol
dehydration unit design capacity (in terms of natural gas flow
rate to the unit per day) is not relevant. According to the
2-181
-------
commenter, keeping records of the design capacity will not show
what the benzene emissions are, and will not really tell anything
about whether the unit qualifies for the one tpy exemption. The
commenter further stated that if the exemption was claimed based
on actual gas throughput being less than 3 MMscf/d, they could
understand a requirement to keep records of how actual gas input
was determined. However, the commenter stated that keeping
records of design capacity would not document whether they
qualify for exemption since the exemption is based on actual gas
input rather than theoretical gas input.
Response: Maintaining records of glycol dehydration unit
natural gas flow rate is required to document that a source meets
the criteria for the throughput exemption under §§63.764(e), and
proposed 63.1274(b) [now codified under §63.1274(d)]. As
proposed, subparts HH and HHH did not require the owner or
operator to keep records for the benzene emission criteria for an
exemption under §§63.764(e), and 63.1274(b). These criteria are
necessary to document that the source qualifies for an exemption
from control requirements based on benzene emissions, therefore,
§63.774 of final subpart HH and §63.1284(d) of final subpart HHH
contain recordkeeping requirements for the benzene emission
criteria to document that the affected source qualifies for the
exemption. The final rule requires records of benzene emissions
determined either by emissions model or direct measurement for
these sources.
Comment: Several commenters were interested in reducing the
recordkeeping and reporting burden on affected sources.
Commenter IV-D-32 recommended streamlining the recordkeeping
requirements, but did not provide any specific recommendations.
Commenter IV-D-06 requested that the EPA allow a "reduced
recordkeeping" option, for monitoring data. The commenter stated
that the EPA should not impose a blanket requirement to retain
every monitored parameter data point, and that such a requirement
2-182
-------
would impose a large paperwork burden with no environmental
benefit. The commenter stated that a compromise, such as the
"reduced recordkeeping" option, which would allow for the storage
of less data if the monitoring system has special enhancements
would be acceptable. The commenter recommended that the EPA use
the HON subpart G, §63.152(g) as a model. The commenter stated
that this comment also applies to subpart HH.
Commenter IV-D-06 also stated that they support
§63.1285(b)(9) and (c) which say that no paperwork requirements
apply to certain exempt sources. The commenter supported the
EPA's making steps toward reducing unnecessary paperwork burdens.
The commenter stated that the same approach should be implemented
in all MACT standards.
Response: The reduced recordkeeping option recommended by
commenter IV-D-06 allows an owner or operator to retain only
daily averages of monitored parameter data if certain monitoring
device design criteria are met. In addition, these requirements
allow an owner or operator to not retain the daily average for
any operating day when the daily average is below the maximum
parameter limit, or above the minimum parameter limit (as
appropriate), provided 6 months have passed without an excursion.
The EPA does not believe that the reduced recordkeeping option is
appropriate for the oil and natural gas production and natural
gas transmission and storage source categories. First, past
performance does not prevent future exceedances. Secondly, the
365-day averaging period option for condensers installed to
comply with subpart HH and the 30-day averaging period option for
condensers installed to comply with subpart HHH require owners or
operators to retain the daily average data to calculate the
appropriate average. Therefore the EPA has not incorporated the
reduced recordkeeping option into subparts HH and HHH.
2-183
-------
However, the EPA did re-evaluate the recordkeeping
requirements and has reduced the number of data points required
to be measured and recorded for monitoring data. The final rules
require continuous monitoring systems to measure and record data
at least every hour, rather than every 15 minutes. The EPA
believes that this will significantly reduce the amount of data
generated by monitoring devices.
Comment: Commenter IV-D-06 stated that they do not support
a requirement to keep records proving entitlement to an
exemption, which takes away some of the value of the exemption.
The commenter stated that since benzene emissions or gas input
can change with time, the records of a historic determination may
not say anything about current conditions. The commenter
maintained that they claim an exemption at their own peril. The
commenter contended that if, at any time, the benzene emissions
or the annual average gas input exceeds the threshold, the
exemption would no longer apply. The commenter stated that this
comment may also apply to subpart HH.
Response; The EPA does not agree with the commenter. Since
the throughput and benzene emission values are annual averages,
some historical data is necessary. Furthermore, the EPA believes
that the exemption has significant value to justify keeping
records. Therefore, the EPA has not modified subparts HH and HHH
in response to this comment.
Comment: Two commenters were particularly concerned with
the onsite recordkeeping requirements. Commenters IV-D-06 and
IV-D-16 requested that the EPA provide for offsite storage of
records and revise Table 2 to say that §63.10(b)(1) does not
apply. The commenters stated that the requirement in
§63.10(b)(1) to store records on-site is not appropriate, as
facilities in remote locations would not have file storage space.
Commenter IV-D-06 proposed that the records be collected
periodically and taken to a central location (such as the same
2-184
-------
location where compliance personnel, who review the records and
prepare the reports, are located). The commenter recommended
that the EPA add a paragraph to the "recordkeeping" section of
subpart HHH expressly allowing offsite storage if the records are
readily accessible. The commenter stated that if subpart HH also
incorporates §63.10(b)(1), this comment also applies to subpart
HH. Commenter IV-D-16 suggested that by replacing the records
retention requirements with inspection by a regulatory official
followed by compliance testing (if necessary), the problem of
retaining records at gas production facilities would be avoided.
Response; The EPA does not believe that replacing the
record retention requirements with inspections followed by
compliance testing (if necessary) would allow inspectors readily
to determine compliance with these standards, at any source.
However, the EPA agrees that owners and operators should be
allowed to retain some data off-site. Therefore, in response to
these comments, the EPA has modified the record retention
requirements in §§63.774(b) and 63.1284(b) as follows:
• Owner or operator must retain the most recent 12 months
of records onsite or the records must be accessible
from a central location by computer or other means that
provides access within two hours after a request;
• Owner or operator may retain the remaining four years
of records offsite; and
• Owner or operator may maintain records in hard copy or
computer-readable form including, but not limited to,
on paper, microfilm, computer, floppy disk, magnetic
tape, or microfiche.
Comment: Commenters IV-D-08, IV-D-22, and IV-G-03 requested
that the EPA merge several reports into one package to be
delivered to the regulatory agency. According to the commenters,
the merged reporting approach prevents multiple, staggered
reports from being generated and transmitted without providing a
comprehensive picture on compliance status. The commenters
recommended that the EPA:
2-185
-------
• allow for reports to be merged through a title V
mechanism for title V facilities;
• merge redundant reports;
• create a unified list of reports; and
• limit frequency to no more than semiannually.
Commenters IV-D-22 and IV-D-34 recommended that the EPA
amend §§63.774 and 63.775 to allow operators to maintain annual
documentation of State or federally enforceable limits, and to
require reports only as necessary based upon the specific device
or operating limit relied upon for emission control. The
commenters urged the EPA to use the same recordkeeping and
reporting requirements as 40 CFR 63 subpart CC (refinery rule) as
follows:
• allow semiannual or annual reports rather than
reporting events as they occur;
• allow the permitting authority and operator to
determine which information will be reported or simply
documented; and
• eliminate duplication for facilities subject to
multiple requirements.
The commenters suggested that this would ensure compliance while
reducing the burden on the operators and permitting agency.
Response; The EPA recognizes that unnecessary monitoring,
recordkeeping, and reporting requirements would burden both the
source and the enforcement agencies. As stated in an earlier
response at the beginning of section 2.11 of this document, the
final rules contain only monitoring, recordkeeping, and reporting
requirements that are necessary to demonstrate compliance.
State and local agencies have the option of enforcing
different, but equivalent, monitoring, recordkeeping, and
reporting requirements if they submit information on their
program to the EPA for approval under the procedures for
delegation of NESHAP authority under section 112(1) of the CAA.
Furthermore, in cases where reporting requirements of State
or local rules duplicate those of subparts HH or HHH, a source
2-186
-------
can work with their State or local title V permit authority to
avoid duplicate submittals.
In response to the coxnmenters' request to limit the
reporting frequency, table 2 of the final rule has been changed
to indicate that the requirement specified in §63.10(e)(3)(i)(C),
for quarterly reporting in cases where monitoring parameters are
out of range or monitors are not operating more than a specified
percent of the time, does not apply. Instead, semiannual
reporting is required for all facilities. As proposed,
facilities were required to report semiannually, but if the
source experienced excess emissions, quarterly reports would be
required.
This change was made because the EPA agrees that the
quarterly reporting system proposed added complexity to the rule,
it may not be helpful for enforcement, and that penalties for
noncompliance are a sufficient disincentive for poor performance.
Further, semiannual reporting is consistent with title V
operating permit reporting requirements. Requiring separate
quarterly reports for some facilities adds complexity and
increases the reporting burden for both the facility and the
enforcement agency. Semiannual reports will provide the
regulatory agency information on excess emissions within about
six months of the occurrence. This is well within the 1-year
period in which the agency can take administrative enforcement
actions as specified under section 113(d) of the CAA.
Comment: Conunenters IV-D-07 and IV-G-03 requested an
adjustment for the reporting requirements for unmanned
facilities, as the reporting requirements seemed unnecessarily
burdensome. Commenter IV-D-07 requested monthly recordkeeping
and annual reporting.
2-187
-------
Response; As stated in the previous response, sources can
discuss with the implementing agency the possibility of
submitting different reports, providing these reports are
equivalent.
Annual reporting was not selected as requested by the
commenters, because it would significantly reduce the EPA's
ability to take administrative enforcement actions. Section
113(d) of the CAA limits the assessment of administrative
penalties to violations that occur no more than 12 months prior
to the initiation of the administrative proceeding. Periodic
reports are a primary means of identifying possible violations,
and annual submittal would not give the enforcement agency time
to review the report and take action on a violation that occurred
early in the reporting period within one year after the event.
Comment: Commenters IV-D-08, IV-D-22, and IV-D-26 stated
that requiring initial notifications within one year is not
realistic. The commenters referred to the Gasoline Distribution
Terminals and Pipeline Breakout Stations standard (subpart R)
which comprises many different sized facilities. According to
the commenters, the initial assessments have taken more than one
year and the EPA has had to amend subpart R to allow the
facilities more time. Therefore the commenters, along with
commenter IV-D-32, recommended the following:
• extend the initial notification period to at least two
years/
• create a tiered notification requirements; and
• allow a single notification to cover multiple
facilities in the same region.
Response; The information required in the initial
notification includes basic facility information, such as name
and address, physical location, identification of the standard
that is the basis of the notification, a facility description,
and a statement of whether the source is a major or area source.
2-188
-------
The EPA believes that all of this information, except the major
or area source determination, should be readily available.
Facilities that are potentially subject to subparts HH and HHH
have been aware that they might have to perform a major source
determination since the date of the proposed rule (February 6,
1998) . Furthermore, by the time the initial notification is due
(assuming an effective date of Hay 15, 1999), these facilities
will have had more than 27 months from the proposal date to
determine their major source status. The EPA believes that there
is ample time to make this determination.
However, the EPA has modified §§63.775 and 63.1285 for
affected sources that are major on or before the date the initial
notification is due (one year after the effective date of the
regulation), that plan to become area sources by the compliance
date. The final rule states that these affected sources that are
major sources but plan to become area sources must include in the
initial notification, a brief, nonbinding description of a
schedule for the actions that are planned to achieve area source
status.
Nothing in subparts HH and HHH prevents single notifications
for multiple facilities, provided the required information for
all facilities is contained in the notifications.
Comment: Comraenters IV-D-08 and IV-D-22 were concerned that
the wording in §63.9(h) might be misunderstood, as far as when
the notification of compliance status is required, or how soon it
has to occur after promulgation. The commenters recommended the
following:
• establish a certain time for Notification of Compliance
Status -- 180 days after promulgation, as provided for
in §63.9(h) (2) (ii)/ and
• reference specific sections in subpart A for the
required contents of this notification (i.e.,
§63.9(h)(2)(i)).
2-189
-------
Response; The EPA agrees that it would be more clear if the
due dates for the notification of compliance status were stated
explicitly in the rule. In addition, the EPA has included
several sections from the General Provisions (40 CFR part 63,
subpart A) directly in subparts HH and HHH. Sections 63.775 (d)
of final subpart HH and §63.1285(d) of final subpart HHH contain
the Notification of Compliance Status Report requirements.
Comment: Commenter IV-D-01 requested that the reporting of
the number of components monitored be a requirement of semiannual
reports. The commenter noted that the reporting requirements in
§61.247(b), for each piece of ancillary equipment subject to
§63.769, do not require the number of components monitored per
reporting period to be included in each semiannual report. The
commenter stated that the number of components is required in 40
CFR 63 Subpart H. The commenter cited the benefits for including
the number of components:
1. Facilitates more rapid review of the report;
2. Provides verification of percent leakers; and
3. Ensures that the facility continues to monitor all
components.
Response; The EPA agrees with the commenter that the number
of monitored components is important to verify the calculation of
the percent leakers and to ensure that a facility continues to
monitor all components. Therefore, §63.775(d)(3) of the final
rule contains requirements for owners and operators subject to
§63.769 to submit the information required under §61.247(a) (of
40 CFR, part 61, subpart V), except for the following:
• The initial report required in §61.247(a) must be
submitted as a part of the Notification of Compliance
Status Report required under subpart HH.
• The number of each equipment (e.g., valves, pumps,
etc.) must be included in the Notification of
Compliance Status Report.
2-190
-------
• Changes in the information submitted in the
Notification of Compliance Status Report must be
submitted in subsequent Periodic Reports.
Comment: Commenter IV-D-01 requested that the EPA delete
the reporting requirement exemption [§63.775(b)(9)] for sources
that are not subject to the control requirements for glycol
dehydration unit process vents. The commenter stated that these
sources may still be subject to the storage vessel and equipment
leak provisions and they must also conduct performance tests and
develop startup, shutdown and malfunction plans. The commenter
further stated that since these facilities are not exempt from
recordkeeping requirements, reporting would "hardly constitute an
excessive burden."
Response; The intent of the reporting requirements
contained in proposed §63.775(b)(9) of subpart HH and
§63.1285(b)(9) of subpart HHH was not to exempt entire
facilities, but to exempt glycol dehydration units that are not
subject to the control requirements, as specified in proposed
§§63.764(e) and 63.1274(b) [now codified at §63.764(e) (1) of
subpart HH and §63.1274(d) of subpart HHH], from the reporting
requirements. To clarify this intent, the final rules specify
that only the units exempt from the control requirements are
exempt frqm the reporting requirements in subpart HH and HHH
[codified at §63.775(b)(7) of subpart HH and §63.1285(b)(7) of
subpart HHH]. Thus, the reporting requirement exemptions do not
apply to other units within the facility that are subject to the
rule, including glycol dehydration units, storage vessels, and
ancillary equipment.
Similarly, §63.764(e)(3) of final subpart HH specifies that
ancillary equipment and compressors that contain or contact fluid
with a VHAP concentration less than 10 percent and that are in
VHAP service less than 300 hours per year are exempted from the
2-191
-------
control requirements of §63.769. Therefore, S63.775(b)(8) of the
final rule contains an exemption from the reporting requirements
for ancillary equipment and compressors that are not subject to
§63.769.
Additionally, it should be noted that although storage
vessels with the potential for flash emissions (as defined in
§63.761) are not subject to the recordkeeping and reporting
requirements, an owner or operator would be required to maintain
records for these sources under §63.10(b)(3) of 40 CFR part 63,
subpart A, which contains recordkeeping requirements for
applicability determinations.
Comment: Commenter IV-D-06 stated that §63.1285(b)(7) does
not specify a deadline for reports on equipment that was
initially exempt, but becomes subject to control requirements due
to process changes. The commenter requested reasonable deadlines
for these reports. The commenter stated that this comment may
also apply to subpart HH.
Response; The EPA agrees that provisions are necessary for
area sources that become major sources due to increases in HAP
emissions or increases in PTE. Therefore, the final rules
specify the following:
• compliance dates for area sources that become major
sources [§63.760(f) of subpart HH and §63.1270(d) of
subpart HHH],
• initial notification requirements for area sources that
become major sources which are correlated to the
compliance dates [§63.775(b)(1) of subpart HH and
§63.1285(b)(1) of subpart HHH], and
• notifications of process changes due within 180 days
after the process change or by the next Periodic
Report, whichever is sooner [§63.775 (f) of subpart HH
and §63.1285(f) of subpart HHH}.
Comment: Commenter IV-D-16 recommended that startup,
shutdown and malfunction reports as required under §63.10(d)(5)
2-192
-------
should not be required under subpart HH. The conunenter did
suggest that if they are required, the startup, shutdown, and
malfunction reporting requirements should only apply to gas
plants, and should be part of the semiannual reports required by
title V rather than a stand alone report.
Response; The startup, shutdown, and malfunction report
enables the EPA to keep track of excess emissions and harm to the
environment, and is only required if the owner or operator does
not follow their startup, shutdown, and malfunction plan.
Therefore, the EPA believes that these reports must be submitted
within seven days of the malfunction. If the owner or operator
is thorough in developing their facility's startup, shutdown, «r»^
malfunction plan, and ensures that the plan is followed during
startup, shutdown, and malfunction events, these reports will not
be necessary. However, it should be noted that the final rule
states that separate startup, shutdown, and malfunction reports
are not necessary if the required information is included in the
Periodic Report for the facility [codified at §63.775(b)(6) of
subpart HH and §63.1285(b)(6) of subpart HHH].
2-193
-------
2 .12 TEST METHODS
Comment: Commenter IV-D-06 stated that Methods 1 or 1A will
usually not be appropriate as required in §63.1282(d)(I).
According to the commenter, Method 1 is only for large stacks and
Method 1A is for smaller stacks, and continual references
throughout Method 1A say the method is for particulate. The
commenter stated that they would not generally be dealing with
particulate under subpart HHH. The commenter recommended that
§63.1282(d)(1) be revised to say that "if we use Method 1A, any
references to particulate do not apply." The commenter stated
that this comment may also apply to subpart HH.
Response; The procedures outlined in Methods 1 and 1A are
to be used only to select a sampling site. However, to clarify
the EPA's intent, the final rules state that Method 1 or 1A of 40
CFR part 60, appendix A must be used to select the sampling site,
and any references to particulate mentioned in Methods 1 and 1A
do not apply [codified at §63.772(e)(3)(i) of subpart HH and
§63.1282(d)(3)(i) of subpart HHH].
Comment: Commenter IV-D-06 stated that the EPA should
revise §63.1283(c)(3) to require Method 21 "with the differences
specified in §63.1282(b) of this subpart." The commenter stated
that it was confusing that §63.1282(b) specified a modified
Method 21 and §63.1283(c)(3) specified "straight" Method 21. The
commenter stated that according to §63.1282(b), something like
modified Method 21 should be used. The commenter stated that
this comment may also apply to subpart HH.
Response; The EPA agrees with this recommendation and has
modified the closed-vent system requirements [now codified at
§63.773(c)(2)(i) and (ii) of subpart HH and §63.1283(c)(2)(i) and
(ii) of subpart HHH] to reference the no detectable emissions
procedure under §63.772(b) of subpart HH and §63.1282(b) of
subpart HHH.
2-194
-------
Comment: Commenter IV-D-09 was concerned with the emissions
test procedures requirement to use Method 18 to determine
concentrations and emissions of HAP and TOC. The commenter
stated that although Method 18 is appropriate for speciating
organic HAP, it is not appropriate for speciating TOC. The
commenter explained that Method 18 uses gas chromatography (GC)
to separate a variety of organic compounds. According to the
commenter, individual peaks could vary widely in resolution,
depending on the complexity of the sample. The commenter further
explained that the error associated with determining TOC by GC
could be substantial as it would reflect the sum of several
errors associated with the separation and detection of individual
organic compounds. Furthermore, the commenter expressed doubt
that any one chromatographic column and set of operating
conditions will cleanly separate the complex organic matrix
present in dehydration units.
The commenter recommended that the EPA either require use of
Method 25A for determining TOC or allow the use of Method 25A as
an alternative means of determining TOC. According to the
commenter, the analytical error associated with discrepancies
between the response factor of the calibration gases used in
Method 25A and the average response factor of the organic gas
matrix will not be as significant as the errors associated with
Method 18. The commenter further recommended that alternate
calibration schemes could be used to ensure safety, if the EPA is
concerned with FID response factors. Specifically, the commenter
suggested that since the FID response factor in Method 25A is
retarded by the presence of oxygenated or chlorinated compounds,
the EPA could specify that the FID at the outlet of a combustion
control device be calibrated with a mixture of methanol in air.
According to the commenter, "since methanol has a relatively high
oxygen-to-carbon ratio, it will have among the most markedly
dampened response factors of any organic compound likely to be
present in the air stream."
The commenter stated that allowing the use of Method 25A
would reduce the cost associated with emissions testing.
2-195
-------
According to the commenter, Method 18 costs approximately two to
three times more than Method 25A because Method 18 requires the
use of more sophisticated laboratory techniques. Additionally,
the commenter noted that since Method 18 does not result in more
accurate TOC emissions estimates, there is not an environmental
benefit associated with an increased cost.
Response; The EPA agrees with the commenter. The aromatic
compounds contained in the vent stream make method 25A an
appropriate method for measuring TOC. The final rules allow the
owner or operator the option of using either method 18 or method
25A [codified at §§63.772(e) (3) (iii) and 63.772(e) (3) (iv) of
subpart HH and §§63.1282(d)(3)(iii) and 63.1282(d)(3)(iv) of
subpart HHH].
Comment: Commenter IV-D-07 requested that the EPA specify
how emissions from the combustion source should be measured.
According to the commenter, Method 25 does not differentiate
between methane, ethane, and VOC, and furthermore, testing grab
samples with a GC/MS tends to be inaccurate. The commenter also
asked whether the 20 ppmv concentration limit is measured "as
propane" or "as methane."
Response; Proposed §§63.772(e)(3) and (4) of subpart HH and
§§63.1282(d)(3) and (4) of subpart HHH [now codified at
§§63.772(e)(3)(iii) and (iv) of final subpart HH and
§§63.1282(d)(3)(iii) and (iv) of final subpart HHH] state that
emissions from combustion sources must be measured using Method
18, 40 CFR part 60, appendix A, or any other method or data
validated according to the applicable procedures in Method 301,
40 CFR part 63, appendix A. As mentioned in the previous
response, the EPA has also modified the test method requirements
to allow the owner or operator the option of using Method 25A,
instead of Method 18.
2-196
-------
When using Method 18, the 20-ppxnv concentration limit is
determined as the sum of the compound concentration as measured
by Method 18. Individual compounds are presented as the compound
(e.g., benzene concentration would be presented as 5 ppmv as
benzene) . If using Method 25A, the measurement should be
presented based on the calibration gas used.
Comment: Commenter IV-G-01 remarked that the proposed
NESHAP states that bagging is the only method for measuring
fugitive hydrocarbon emissions. The commenter stated that the
High Volume Collection System has been demonstrated to be as
effective as the bagging method, as shown in EPA-600/R-95-167.
Response; Although the High Volume Collection System could
be an alternative method for measuring fugitive HAP emissions,
the EPA determined that the MACT floor for equipment leaks from
ancillary equipment and compressors in the oil and natural gas
production source category was the level of control required
under 40 CFR part 60, subpart KKK. Therefore, the equipment leak
standards are based on work practices and operational practices
equivalent to those required under subpart KKK, rather than
emission standards.
Comment: Commenter IV-G-15 provided a letter from a
technical consultant to the commenter that describes how EPA
Methods 0030/5040 Volatile Organic Sampling Train (VOST) can be
used with some modification to characterize accurately emissions
from glycol dehydration units. The consultant described the
modified method as the new "Bag VOST" technique, similar to
California Air Resources Board Method 422.
Commenter IV-G-16 and commenter IV-G-15 provided proposed
sampling procedures for glycol dehydration units. According to
the commenters, the proposed procedures would be used to estimate
uncontrolled emissions and control efficiency for units
controlled with condensers, and are based on EPA Methods 1, 2,
3B, 4, and 18 modified. The commenters stated that gas samples
2-197
-------
are collected over a one hour period. The commenters referred to
the method of analysis as GC/FID and GC/TCD.
Commenter IV-D-03 provided for the EPA's information, a copy
of the paper entitled Flash Vaporization Emissions Test Method
for Storage Tank VOC & HAP, which was presented at the October
1997 EPA/A&WMA Emission Inventory Conference in RTF, NC. The
commenter stated that the paper describes the EquiVap™ test
method and its ability to quantify accurately and speciate
storage tank emissions based on fundamental thermodynamic
principles using cost effective conventional labware and software
analytical tools.
Response; In order for the EPA to approve alternative test
methods, the Agency must receive an analysis according to the
procedures in 40 CFR part 60, appendix A, Method 301. Additional
guidance for obtaining EPA approval for alternate test methods
and procedures may be found on the Internet at:
http://www.epa.gov/ttn/emc.
Comment: Commenter IV-G-02 noted that §63.772(a) appears to
be in conflict with §63.769 (a). The commenter explained that
§63.772(a) sets out a complicated procedure using Method 305 or
Method 25D to determine the HAP content of material for
applicability of equipment leak standards and §63.769 (a) requires
the use of the method in §61.245(d) (incorporating ASTM Method
D-2267). The commenter requested a clarification that either
method can be used to give owners and operators maximum
flexibility to choose most economical method available to them.
Response; The EPA agrees that proposed §63.769(a) and
§63.772(a) are inconsistent. The test method specified in
§61.245(d), ASTM Method D-2267, is no longer considered by the
EPA to be a valid test method. Therefore, the EPA has modified
§63.769(a) in the final rule as follows:
"... that contains or contacts, a fluid (liquid or
gas) that has a total VOHAPVHAP. concentration equal to
2-198
-------
or greater than 10 percent by weight (determined
according to the provioiono of 40 OFR
61.245(d)procedures specified in S63.772(al) ..."
In addition, the EPA evaluated the procedures for determining
VHAP concentration for the applicability to the equipment leak
standards under proposed §63.772(a) and determined that the
procedures were not appropriate for the oil and natural gas
production source category. The EPA believes that Method 18 of
40 CFR part 60, appendix A and the procedure specified in 40 CFR
63.180(d) of this part are appropriate for determining the VHAP
concentration of fluid contained in or in contact with ancillary
equipment or compressors. Therefore, §63.772(a) of the final
subpart HH reads as follows:
(a) Determination of material VHAP or HAP
concentration to determine the applicability of the
equipment leak standards under this suboart (§63.769).
Each piece of ancillary equipment and compressors are
presumed to be in VHAP service or in wet gas service
unless an owner or operator demonstrates that the piece
of equipment is not in VHAP service or in wet gas
service.
(1) For a piece of ancillary equipment and
compressors to be considered not in VHAP service, it
must be determined that the percent VHAP content can be
reasonably expected never to exceed 10.0 percent by
weight. For the purposes of determining the percent
VHAP content of the process fluid that is contained in
or contacts a piece of ancillary equipment or
compressor, Method 18 of 40 CFR part 60, appendix A,
shall be used.
(2) For a piece of ancillary equipment and
compressors to be considered in wet gas service, it
must be determined that it contains or contacts the
field gas before the extraction of natural gas liquids.
Comment: Commenter IV-D-06 requested that the EPA clarify
"background level" issues, for instrumental leak detection
monitoring. The commenter referred to §63.1282(b)(5), which
requires that the procedures in Method 21 should be used to
2-199
-------
determine background levels. However, the commenter stated that
Method 21 does not have any procedures to determine background
levels. According to the commenter, Method 21 says how to
monitor for concentrations of certain substances. Additionally,
the commenter recommended that the EPA clarify whether or not
there is an option to either determine background level. The
commenter stated that the regulatory burden would be reduced and
the environment would actually be benefitted by not determining a
background level. The commenter suggested that the EPA use the
current version of the HON subpart H, §63.180(c) and (c)(2). The
commenter stated that this comment also applies to subpart HH.
Response; The EPA does not agree with the commenter's
statement that Method 21 does not contain procedures to determine
background emission levels. In fact, section 4.3.2 of EPA Method
21 is a section for determining local ambient concentrations (40
CFR part 60, appendix A, Method 21, §4.3.2), which is the
methodology for determining background emission concentrations.
The EPA agrees that an option for determining background
emissions is appropriate. Therefore, §63.772(c)(5) of final
subpart HH and §63.1282(b)(5) of final subpart HHH provide the
owner or operator the option of not determining background
emissions. However, it should be noted that if an owner or
operator chooses not to determine background emissions, and
Method 21 shows emissions greater than 500 ppmv, the equipment
will not be in compliance.
2-200
-------
2.13 COMPLIANCE
2.13.1 Compliance Procedures
Comment: Commenter IV-D-22 recommended that subpart HH
include a provision allowing for delay of repair in the event the
repairs cannot be made on-line and a unit shutdown will be
required. The commenter recommended a provision similar to the
provision in 40 CFR §63.171(a), referenced in the Refinery MACT
rule [40 CFR §63.648(c)], which allows for delay of repair if the
repairs require a shutdown.
Response; Proposed §63.769(c) states that the owner or
operator of ancillary equipment must "meet the requirements
specified in 40 CFR 61.241 through 61.247 . . . , " including
the provisions contained in §61.242-10, allowing for a delay of
repair if the repair cannot be made on-line and a unit shutdown
would be required. Therefore, the equipment leak standards of
subpart HH already contain the provisions requested by the
commenter.
Comment: Commenters IV-D-18 and IV-D-25 stated that the
procedure specified in §63.772 corrects the measured outlet
concentration of HAP to 3 percent oxygen. According to the
commenters, many thermal and catalytic oxidizers properly operate
with oxygen levels in the exhaust stream near 20 percent and the
correction to 3 percent oxygen would make the concentration-based
limit unnecessarily restrictive. Therefore, commenter IV-D-25
recommended that §63.772 (e) (4) be changed to reflect that the
concentration of TOC shall be corrected to the designed oxygen
content in the outlet stream and the equation in
§63.772(e) (4) (iii) (B) should be modified accordingly.
Response: Section 63.771(d)(1)(i)(B) (control requirements)
provides an option requiring combustion devices to reduce "the
concentration of either TOC or total HAP in the exhaust gases at
the outlet to the device to a level equal to or less than 20
parts per million by volume, on a dry basis, corrected to 3
2-201
-------
percent oxygen . . . . " To make a direct comparison between the
enclosed combustion device total HAP concentration limit
specified in §63.771(d)(1)(i)(B), and the TOG or total HAP
emissions measured using the procedures specified in §63.772(e),
a correction to 3 percent oxygen is necessary. Without this
correction, a direct comparison of measured emissions to the
concentration limit would not be possible. Therefore, the EPA
does not see any reason to modify §63.772 (e) (4) or the equation
contained in proposed §63.772(e)(4)(iii)(B) [now codified at
§63.772 (e) (3) (iv) (C) (2.) of the final rule] in response to this
comment.
Comment: Commenter IV-D-35 requested that the alternative
condenser evaluation allowed in §63.1282(e) specify that the
liquid streams to be sampled are before and after the condenser,
to avoid any confusion regarding the testing before and after the
still or reboiler. [Note: this comment also applies to
§63.772(f)] .
Response; The commenter's request that proposed §63.1282(e)
specify that the liquid sample streams are "before and after the
condenser" is incorrect. The GRI report entitled "Atmospheric
Rich/Lean Method for Determining Glycol Dehydrator Emissions"
(GRI-95/0368.1) specifies the following sample point locations
for collecting rich and lean glycol samples:
1. Rich Glycol: select a sample point between the glycol
pump and the reboiler.
2. Lean Glycol: select a sample point between the
reboiler and the contactor, if a charcoal filter is not
in line between the reboiler and the contactor, OR
between the reboiler and the charcoal filter, if a
charcoal filter is between the reboiler and the
contactor.
Therefore, the sample locations are required to be before and
after the reboiler, not before and after the condenser.
2-202
-------
The EPA's intent was for the ARL methodology specified in
the 6RI report to be used in conjunction with GLYCalc to
determine condenser performance. To clarify this intent,
proposed §§63.772(f) and 63.1282(e) [now codified at
§63.772(e)(5) of final subpart HH and §63.1282(d)(5) of final
subpart HHH] have been modified to specify that the ARL method
can be used to provide inputs for use in conjunction with
GLYCalc.
Comment: Commenter IV-G-07 supplied a detailed
"engineering assessment" methodology for calculating annual HAP
and VOC emissions from condenser-controlled dehydration units
(for possible use in emissions reporting). This method would
use: (1) GLYCalc to estimate uncontrolled emissions, (2) a
process design model (e.g., Hysim, Aspen, PD+, etc.) to estimate
emissions as a function of condenser outlet temperature, (3) a
curve of emissions vs. ambient air temperature, constructed based
on a condenser design that produces an outlet vapor/liquid stream
in equilibrium at least ten degrees Fahrenheit above ambient
temperature, and (4) National Climatic Data Center temperature
data for the station nearest the site to construct an annual dry
bulb temperature histogram. Annual emissions would be calculated
by integrating the emissions vs. air temperature curve over the
temperature histogram.
Commenters IV-G-15 and IV-D-16 described models that exist
for estimating emissions from glycol dehydration units.
According to the commenters, GLYCalc, written by Radian
Corporation, and PROSIN, written by Bryan Research and
Engineering, are approved by the TNRCC. A third model written by
OPC DRIZO, is used for design purposes by OPC, but is not
available to the public. The consultant made the statement that
the emissions models used generally underestimate the actual
emissions generated, and that nothing beats a good stack test.
Response; The EPA reviewed the information supplied by the
commenters. The EPA evaluated GLYCalc from GRI as a tool for
2-203
-------
estimating emissions from glycol dehydration units24 and the EPA
recommends the use of this model for the development of emission
inventories to meet Clean Air Act requirements. The EPA
understands that the program may overestimate emissions from
these units, but believes that the use of accurate input data
will reduce this potential.
Proposed §§63.771(d)(3)(iv) and 63.1281(d)(3)(iv) [now
codified at §63.772(e)(4) of final subpart HH and §63.1272(d)(4)
of final subpart HHH] contain specific requirements for design
analyses which may be used to demonstrate compliance with the
control device performance requirements. The commenters'
suggested approaches are acceptable, provided they comply with
the requirements for condenser design analyses as specified in
final §63.772(e) (4) (i)(D) or §63.1282(d)(4)(i) (D). It should be
noted that any disagreements between the owner or operator and
the Administrator would be resolved by conducting a performance
test as specified in final §§63.772(e)(4)(ii) or
63.772(d)(4)(ii). The EPA has not added guidance for estimating
emissions, for reporting, to the final rules. However, the EPA
will publish implementation guidance following promulgation of
subparts HH and HHH.
Comment: Commenter IV-D-07 interpreted §63.1282(d) (1) (i) (A)
to mean that the sampling site must be located upstream of the
control device. The commenter also suggested that the EPA
include a definition for the term final product recovery device.
Response; The control device inlet sampling site must be
located at the inlet of the first control device. To clarify,
24Memorandum from Jones, L.G., U.S. EPA/Emissions
Measurement Branch, to J.D. Mobely, U.S. EPA/ Emission Factor and
Inventory Group. "Glycol Dehydrator Emissions Test Report and
Emissions Estimation Methodology." April 13, 1995.
2-204
-------
the EPA has modified proposed §§63.772(e)(1)(i)(A) and
63.1282(d)(1)(i)(A) [now codified at §§63.772(e)(3)(i)(A) and
63.1282(d)(3)(i)(A)] as follows:
(A) ... inlet sampling sites shall be located at
the inlet of the first control device and at the outlet
of the final control device.
A recovery device is one used for recovering chemicals for
fuel value, use, reuse, or for sale for fuel value, use, or
reuse/ and is not considered a control device. This definition
of recovery device is imbedded within the definition of control
device.
2.13.2 Compliance Determination
Comment: Commenter IV-D-06 recommended that the EPA provide
an exemption from performance testing for RCRA-regulated
hazardous waste incinerators. According to the commenter,
RCRA-regulated hazardous waste incinerators have already had to
demonstrate compliance with very stringent emission standards
under RCRA, and no further compliance demonstration is needed for
MACT standards. The commenter requested that the EPA use the
wording in §63.116(b)(5) of subpart G of the HON, verbatim, in
§63.1281(d)(3)(i) as a new paragraph (d)(3)(i)(E). The commenter
stated that this comment also applies to subpart HH.
Response; The EPA is not aware of any oil and natural gas
exploration and production or natural gas transmission and
storage facilities that would have RCRA industrial furnaces.
However, the EPA has added the recommended language based on
§63.116(b) (5) of subpart 6 of the HON to final subparts HH and
HHH [codified at §63.772(e)(1)(v) of final subpart HH and
§63.1282(d) (3)(v) of final subpart HHH].
Comment: Commenter IV-D-06 suggested that the EPA include
an exemption from performance testing under subpart HHH, for
control devices that have already had a performance test under
other EPA regulations. The commenter stated that the EPA should
2-205
-------
use the language in §63.116(b)(3) of the HON, verbatim, and used
in subpart HHH as a new paragraph (d)(3)(i)(F). The commenter
stated that this comment also applies to subpart HH. [Note: The
commenter did not provide section number in subpart HHH, it can
be assumed that they were referring to §63.1281.)
Response; The EPA agrees that control devices that have had
a performance test under other EPA regulations should be exempt
from the performance test requirements under subparts HH and HHH.
Therefore, the EPA has added a new paragraph (vi) to
§63.772(e)(1) of final subpart HH and §63.1282(d)(1) of final
subpart HHH as follows:
(vi) A control device for which a performance test
was conducted for determining compliance with a
regulation promulgated by the EPA and the test was
conducted using the same methods specified in this
section and either no process changes have been made
since the test, or the owner or operator can
demonstrate that the results of the performance test,
with or without adjustments, reliably demonstrate
compliance despite process changes.
Comment: Commenter IV-D-06 suggested that subpart HHH
specify what constitutes a compliance demonstration for flares.
According to the commenter, subpart HHH never says whether a
compliance demonstration is required for flares, or how to
conduct a compliance demonstration. The commenter cited three
examples where §63.1281(d) (3) (ii) points to §63.11(b) of subpart
A, and does not contain a specific requirement for a compliance
demonstration:
• Section 63.11(b) requires the fuel for a flare to have
a certain minimum net heating value and how the net
heating value would be determined. However, nothing in
§63.11(b) says the owner or operator must actually
perform the calculations at a certain time, on a
certain fuel, to demonstrate compliance.
• Section 63.11(b) requires that flares be designed for,
and operated with, no visible emissions and what test
method must be used to determine whether there are
visible emissions. However, nothing in §63.11(b) says
2-206
-------
the owner or operator must actually perform the visible
emissions test at a certain time, to demonstrate
compliance.
• Section 63.11(b) requires that steam-assisted and
nonassisted flares be designed for and operated with an
exit velocity less than a certain figure, with
specified exceptions. Air-assisted flares are given a
different limit for the exit velocity. The regulations
also provide a method for determining the exit velocity
in each case. However, nothing in §63.11(b) says the
owner or operator must determine the exit velocity at a
certain time, to demonstrate compliance.
The commenter contended that if subpart HHH references §63.11(b)
for flares, nothing requires a compliance demonstration for
flares. The commenter stated that they believe that enforcement
actions have been taken by some EPA personnel, based on the
interpretation that §63.11(b) requires a compliance demonstration
for flares. However, the commenter does not think that §63.11{b)
actually requires a compliance demonstration. Therefore, the
commenter requested that subpart HHH specify what elements are
included in the compliance demonstration and what the deadline
is. The commenter provided the following language that could be
inserted to replace §63.1281(d)(3)(ii).
(ii) Notwithstanding any other provision of this
subpart, if an owner or operator uses a flare to comply
with any of the requirements of this subpart, the owner
or operator shall comply with (ii)(A) through (ii)(C)
of this paragraph. The owner or operator is not
required to conduct a performance test to determine the
percent emission reduction or outlet HAP or TOC
concentration. If a compliance demonstration has been
conducted previously for a flare, using the techniques
specified in (ii)(A) through (ii)(C) of this paragraph,
that compliance demonstration may be used to satisfy
the requirements of this paragraph if either no
deliberate process changes have been made since the
compliance demonstration, or the owner or operator can
demonstrate that the results of the compliance
demonstration reliably demonstrate compliance.
(A) Conduct a visible emissions test using the
techniques specified in §63.11(b)(4) of subpart A;
(B) Determine the net heating value of the gas
being combusted, using the techniques specified in
§63.ll(b)(6) of subpart A; and
(C) Determine the exit velocity using the
techniques specified in either §63.11(b)(7)(i) and
2-207
-------
§63.1Kb) (7) (iii) where applicable, or §63 .11 (b) (8) , as
appropriate.
The commenter stated that this comment also applies to subpart
HH.
Response: The EPA agrees that specific requirements for a
flare compliance test are necessary. Therefore, the final rules
contains §63.772(e)(2) (for subpart HH) and §63.1282(d)(2) (for
subpart HHH) as follows:
(2) An owner or operator shall design and operate
demonstrate the performance of each flare in accordance
with the requirements specified in §63.1Kb) and in
paragraphs (e)(2)(i) and (ii) [or (d)(2)(i) and (ii)
for subpart HHH] of this section.
(i) The compliance determination shall be
conducted using Method 22 of 40 CFR part 60, appendix
A, to determine visible emissions.
(ii) An owner or operator is not required to
conduct a performance test to determine percent
emission reduction or outlet organic HAP or TOC
concentration when a flare is used.
Comment: Regarding the requirement for combustion sources
(95-percent emission reduction, or 20-ppmv outlet concentration),
commenter IV-D-07 asked how compliance with the requirement is to
be demonstrated, and under what operating conditions the
requirement is applicable.
Response; Compliance with the requirement for combustion
sources (95-percent emission reduction, or 20-ppmv outlet
concentration) is demonstrated by monitoring specified operating
parameters [see §63.773(d)(3)(i)(A) through (G) of final
subpart HH or §63.1283(d)(3)(i)(A) through (G) of final subpart
HHH for specific parameters] . For each operating parameter
monitored, the owner or operator selects a minimum or maximum
operating parameter value (as appropriate) at which the control
device must be operated continuously to achieve the applicable
performance requirements. The minimum or maximum operating
2-208
-------
parameters are established based on the control device
manufacturer's recommendation along with either values measured
during a performance test or the control device design analysis.
For example, §63.773(d)(3)(i)(A) of final subpart HH requires an
incinerator to be equipped with a temperature monitoring device.
The owner or operator must establish a minimum operating
temperature, either by performance test, or design analysis, at
which the incinerator must be operated continuously to achieve
95-percent emission reduction. The data collected by the
temperature monitoring device must not fall below the minimum
operating temperature.
Emissions from a glycol dehydration unit are required to be
routed to a control device if the glycol dehydration unit has a
natural gas throughput greater than 3 MMscf/d or benzene
emissions greater than I tpy. If the owner or operator chooses
to use a combustion device to comply with the process vent
control requirements, the combustion device must achieve
95-percent emission reduction or 20-ppmv outlet concentration.
In addition, the final rule allows owners or operators to install
control technologies that reduce emissions from glycol
dehydration unit process vents to 1 tpy of benzene or less.
Comment: Commenter IV-D-14 asked how operating parameter
values used in demonstrating compliance would be assigned and who
would be responsible for determining the appropriate values.
Response; The owner or operator is responsible for
determining the appropriate operating parameter values (either as
a minimum or a maximum) to use to demonstrate compliance
[§63.773(d)(5) of final subpart HH and §63.1283(d)(5) of final
subpart HHH]. The operating parameter values must be based on
either performance testing or design analysis, supplemented with
2-209
-------
the manufacturer's recommendations [§63.773(d) (5) (i) (A) and (B)
of final subpart HH and §63.1283(d)(5) (i) (A) and (B) of final
subpart HHH]. The established operating parameters and the
rationale for choosing them, must then be submitted in the
Notification of Compliance Status Report [§§63.775(d)(5) and
63.1285(d)(5) of the final rules].
Comment: Comtnenter IV-D-12 requested that the EPA clarify
the methods proposed for demonstrating adequacy of control
performance and that the EPA provide examples to show how these
methods should be applied.
Response; The EPA agrees that subparts HH and HHH should be
clarified to specify when performance tests or design analyses
should be conducted. Therefore, the EPA has language to subparts
HH and HHH to clarify how compliance is demonstrated using
performance tests and design analyses. Section 63.772(f) of
final subpart HH and §63.1282(e) of final subpart HHH contain
requirements for demonstrating compliance for all control
devices. As an alternative, the owner or operator may choose to
demonstrate compliance using condensers in accordance with
§63.772(g) of final subpart HH or §63.1282(f) of final
subpart HHH.
Sections 63.772(e)(4) and 63.1282(d)(4) of the final rules
specify that the performance tests must be conducted according to
the schedule in §63.7(a)(2) of subpart A and the results must be
submitted in the Notification of Compliance Status Report
required in the appropriate subpart. Examples of how test
methods should be applied will be included in implementation
guidance to be published after the promulgation dates for
subparts HH and HHH.
Comment: Commenter IV-D-12 stated that a design analysis is
the appropriate way to demonstrate control device efficiency
2-210
-------
because methods to perform emissions performance tests on
dehydrator vent streams are unreliable. However, the commenter
stated that approval of the design analysis on a case-by-case
basis is inappropriate and could result in "an administrative
bottleneck to efficient implementation of the standard." The
commenter recommended that the EPA define calculations and
records to demonstrate control efficiency and that records be
kept on file similar to that required by some NSPS.
Response; The EPA provided the option for an owner or
operator to perform design analyses instead of a performance test
because design analyses are a generally less expensive option,
and simpler to perform than conducting a performance test.
However, due to the potential for variability in design analyses,
the EPA believes that approval on a case-by-case basis is
necessary. Therefore, the EPA has not modified subparts HH and
HHH in response to this comment.
Comment: Commenter IV-D-05 referred to §63.772 (f) which
allows for an alternative to the documented test procedures of
§63.772(e) if the ARL test is performed in accordance with the
procedures in GRI-95/0368.1. According to the commenter, this
type of test will not give results of how a condensing unit is
working because glycol concentrations are not changed just
because a condensing unit is added to the still column vent. The
commenter recommended that the EPA allow the use of GLYCalc as an
alternative since it was designed to model the emission effects
of various control devices. In support of this recommendation,
the commenter noted that using GLYCalc is allowed to determine
exemptions from the control requirements.
Commenter IV-D-22 recommended that the EPA amend Section
63.772(b) (2) by adding subpart (iii) as follows:
(iii) The owner or operator shall determine an
average emissions rate of benzene in tons per year
following the procedures specified in GRI Publication
95/0368, March 1996.
2-211
-------
The commenter stated that the ARL method has been approved as an
accurate method of measuring emissions by the EPA Emission Factor
and Inventory Group (memorandum, dated October 26, 1995). The
commenter also recommended that the EPA amend §63.772(f) to make
it clear that the ARL method can be used to calculate BTEX
emissions from a dehydrator in an uncontrolled state. To
calculate condenser control efficiency, as intended in
§63.772(f), the results of the ARL method must be used in
conjunction with GLYCalc to calculate condenser control
efficiency. The commenter recommended that the EPA amend
§63.772(f) by inserting "in conjunction with GLYCalc to calculate
control device performance" at the end of the proposed section.
Response; The ARL method is intended to be used for
determining uncontrolled emissions from the glycol dehydrator.
The EPA intended that the ARL method would be used in conjunction
with GLYCalc to determine condenser performance. Therefore,
proposed §§63.772(f) and 63.1282(e) [now codified at
§63.772(e)(5) and 63.1282(d)(5) of the final rules] have been
modified to clarify that the results from the ARL method can be
used as inputs to GLYCalc (Version 3.0) to estimate condenser
control efficiency.
Additionally, the EPA agrees with the commenter that the ARL
method should be allowed as an input to GLYCalc for estimating
actual benzene emissions. Therefore, §§63.772(b)(2)(i) and
63.1282(a)(2)(i) of the final rules state that the procedures
documented in the GRI report entitled, "Atmospheric Rich/Lean
Method for Determining Glycol Dehydrator Emissions"
(GRI-95/0368.1), can be used in conjunction with GLYCalc to
determine actual benzene emissions.
Comment: Commenter IV-D-07 questioned whether the
parameters listed in §63.1281 for the condenser design analysis
(i.e., vent stream composition, constituent concentrations, flow
2-212
-------
rate, relative humidity, and temperature) refer to the natural
gas stream entering the contactor tower. For the outlet, the
commenter also asked if they need to determine the outlet organic
concentration levels or levels of controlled emissions. The
commenter noted that not all condensers have a specific coolant
fluid for which inlet and outlet temperatures can be measured
(e.g., air-cooled vs. glycol-cooled, water-cooled, or forced
draft).
Response; The parameters required for the condenser design
analysis (i.e., vent stream composition, constituent
concentrations, flow rate, relative humidity, and temperature)
refer to characteristics of the emission stream to be treated by
the condenser, in this case, glycol dehydration unit process vent
streams (consisting of reboiler vent emissions, or flash tank
vent emissions, or both).
The design analysis was intended to provide some relief from
the burden of demonstrating compliance with the control device
performance requirements, by giving the owner or operator an
alternative to performance testing. The parameters required for
the design analysis (i.e., design outlet organic concentration,
design average temperature of the condenser exhaust vent stream,
and the design average temperatures of the coolant fluid at the
condenser inlet and outlet) are required to show that the
condenser was designed, based on the process vent stream
characteristics, to achieve an emission reduction of 95 percent.
The intent of these design analysis requirements was not to
require monitoring of outlet organic concentration level and
coolant fluid inlet and outlet temperature. However, the design
analysis should show that these factors were considered in the
design analysis when establishing the operating temperature of
the condenser exhaust stream, which is required to be monitored.
2-213
-------
The EPA does not see any reason to modify subparts HH and HHH in
response to this comment.
Comment: Commenters IV-D-08, IV-D-22, and IV-D-30
recommended that the EPA specify GLYCalc as an allowed tool to
perform the design analysis for a condenser to determine
compliance with the control requirements. Commenters IV-D-08 and
IV-D-22 suggested that the design condenser exhaust stream
temperature should be established as the maximum temperature that
allows the condenser to achieve the control required by
§63.771(d) (1) (ii) . Commenter IV-D-22 stated that if GLYCalc is
used to perform the design analysis, the EPA should delete the
references to relative humidity, (air) temperature, and the
design average temperatures of the coolant fluid at the condenser
inlet and outlet, since these parameters are not required by
GLYCalc. Commenter IV-D-30 recommended that the EPA add the
following language to §63.771(d)(3)(iv)(A)(4J: "For example,
GLYCalc Version 3.0 or higher is an acceptable design analysis
tool for the purposes of this paragraph."
Response; The EPA agrees that it is appropriate to allow an
owner or operator to use GLYCalc as an alternative to the design
analysis for condensers specified in proposed
Si63.771(d) (3) (iv) (A) (1) and 63.1281(d) (3) (iv) (A) (4) [now
codified at §§63.772(e)(4)(i)(D) and 63.1281(d)(4)(i)(D) of the
final rules] and has modified §§63.772(e)(4)(i)(D) and
63.1282(d)(4)(i)(D) of the final rules accordingly. The
requirements for using relative humidity, temperature, and design
average temperatures of the coolant fluid at the condenser inlet
and outlet in the design analysis are necessary to determine
condenser performance when GLYCalc is not used. Therefore, the
EPA has retained these requirements.
Comment: Commenter IV-G-09 stated that the direct benzene
test in §63.1282(a)(2)(ii) is subject to significant inaccuracy
because of the low pressure differential available for sampling
2-214
-------
dehydration unit vents. Therefore, according to the commenter,
the GLYCalc model in §63.1282(a)(2)(i) or some alternative model
is critical to demonstrate compliance successfully.
Response; The EPA agrees with the commenter. As proposed,
subparts HH and HHH provide for the use of GLYCalc as an
alternative for estimating benzene emissions.
Comment: Commenter IV-D-35 referred to proposed
§63.1282(a) (2) (i) which requires the owner or operator to
determine annual benzene emissions using GLYCalc. Inputs to the
model are required to be "representative of actual operating
conditions." The commenter was concerned that the variation in
the ranges of values and the affect on emissions was uncertain.
The commenter requested that §63.1282(2)(i) include some guidance
on how the "representative" values for use with GLYCalc are to be
developed. The commenter stated that they have worked with the
industry to develop an approach with no success.
Similarly, the commenter was concerned that proposed
§63.1282 (a) (2) (ii) does not address how the glycol dehydration
unit should be set up for a benzene emission test. According to
the commenter, the activity level during a test may not represent
yearly operations. The commenter requested guidance to produce
meaningful results and stated that the use of the GLYCalc model
may become very limited without guidance on input parameters in
the proposed regulation.
Response: Because of the great variability in the operation
of glycol dehydration units in the oil and natural gas production
and natural gas transmission and storage source categories, the
EPA could not develop guidance on what may be considered
representative parameters within the final rules. However, the
EPA plans to publish implementation guidance for these rules,
following promulgation. It is up to the owner/operator to select
and document the appropriate values for the parameters that they
will be using as representative values.
2-215
-------
Comment: Commenter IV-G-14 requested that GLYCalc not be
used to determine BTEX emissions from reboilers. The commenter
was concerned that conditions may be altered to slant emissions
to the low side. The commenter also stated that some companies
are taking gas samples downstream of glycol contact towers,
resulting in unrepresentative temperatures of the gas being
processed being reported. The commenter stated that the GLYCalc
program is not as acceptable as actual testing because of
potential inaccuracies. However, the commenter stated that if
typical conditions were listed as a guide for the reviewer to go
by, then GLYCalc would be an acceptable method for determining
emissions. The commenter noted that if the sampling port is
downstream of a glycol contact tower the measurement would not be
accurate because of glycol's affinity for benzene. The commenter
also stated that safety concerns and moisture concerns are not
problems for good stack testers.
Response; As stated in a previous response, the EPA
reviewed GLYCalc and has determined that the program is an
acceptable method for estimating emissions from glycol
dehydration units.25 The EPA also believes that owners /operators
who use this program will determine accurate emission estimates
due to the documentation required to establish representative
parameters. The EPA intends to publish implementation guidance
after the promulgation of subparts HH and HHH to aid inspectors
in the evaluation of compliance with the regulations.
Sections 63.772(b)(2)(i) and 63.1282(a)(2)(i) of the final
rules allow an owner or operator to use the ARL method as an
input to GLYCalc. The ARL method specifies where the gas samples
must be collected. When not using the ARL method, the GLYCalc
25 Reference 24.
2-216
-------
manual26 provides guidance for the best locations for collecting
gas samples. Therefore, §§63.772(b)(2) (i) and 63.1282(a) (2)(i)
of the final rules specify that the owner or operator should use
the procedures specified in the GLYCalc Manual for collecting gas
samples.27
Comment: Commenter IV-G-15 recommended that the Agency
require stack testing of controlled and uncontrolled emissions
points whenever possible, especially where condensers are used.
The commenter submitted examples of testing methods (including
proposed sampling procedures for testing glycol dehydrators
submitted at the commenter's request by commenters IV-G-16 and
IV-G-17), and requested that the test methods in the final rule
specify how to test instead of providing general guidance on
testing. The commenter noted that LDEQ has conducted emissions
tests of compressors and boilers in addition to other emissions
points in the source category. The commenter noted that benzene
emissions are significant and should be tested rather then
modeled.
Response: The EPA provides an owner or operator with the
option of conducting an emission source test (i.e., a stack test)
or design analysis. The EPA realizes that an emission source
test may impose a severe burden on many owners or operators (the
costs associated with planning and conducting such a test) in the
oil and natural gas production and natural gas transmission and
storage source categories. Thus, the EPA included design
analyses as an acceptable option for demonstrating compliance to
reduce the overall burden. The EPA includes references to
26 Radian International LLC. Technical Reference Manual for
GLYCalcTM; A Program for Estimating Emissions from Glycol
Dehydration of Natural Gas, Version 3.0. Prepared for Gas
Research Institute. Chicago, Illinois. GRI-96/0091. March 1996.
pp. 7-1 through 7-14.
27 Reference 26.
2-217
-------
appropriate emission measurement methods and does not believe
that articulating how to test is necessary. Therefore, the EPA
has not modified subparts HH and HHH in response to this comment.
Comment: Commenters IV-D-22 and IV-G-11 requested that the
EPA clarify compliance obligations by expressly indicating in
§63.760 (e) that the facilities failing to meet storage vessel
thresholds of §63.764(c)(2) or that meet the glycol unit
exemptions of §63.764 (e) are not subject to any requirements of
subpart HH.
Response; The EPA has determined that for the class of
storage vessels that do not have the potential for flash
emissions, the MACT floor is no control. Therefore, storage
vessels with the potential for flash emissions are defined in
final subpart HH to mean any storage vessel that contains a
hydrocarbon liquid with a stock tank GOR greater than or equal to
1,750 scf/barrel and an API gravity greater than or equal to 40
degrees, and a hydrocarbon liquid throughput greater than or
equal to 500 bpd. Since the affected source at an oil and
natural gas production facility is defined, in §63.760(b) of the
final rule, as each storage vessel with the potential for flash
emissions, owners or operators of storage vessels that do not
meet the definition of storage vessels with the potential for
flash emissions, have no further obligations for those storage
vessels under subpart HH. However, it should be noted that the
owner or operator is subject to §63.10 (b) (3) of subpart A and
must maintain records of applicability determinations.
Each glycol dehydration unit with an actual annual average
natural gas throughput less than 3 MMscf/d or with actual average
benzene emissions less than 1 tpy are not exempt from the
subpart, but are subject to recordkeeping and reporting
requirements. In addition, these units must be included in the
2-218
-------
calculation of PTE for major source applicability determinations.
Furthermore, the exemptions provided in §63.764(e) do not apply
to entire facilities, but to each individual glycol dehydration
unit. Therefore, providing a general exemption for a facility in
§63.760(e) would not be applicable.
As proposed, ancillary equipment and compressors located at
natural gas processing plants with a VHAP concentration less than
10 percent and in VHAP service for less than 300 hours per year
were not subject to the recordkeeping and reporting requirements.
However, the EPA believes that recordkeeping requirements for
these equipment are appropriate. Therefore, §63.755(d)(2) of
final subpart HH contains recordkeeping requirements for the
documentation of the information and data used to determine which
ancillary equipment and compressors are exempt from the control
requirements of §63.769 of subpart HH.
2.13.3 Compliance Dates
Comment: Commenter IV-D-38 suggested that §63.760 (f) (2) be
modified so that the effective date for new construction and
reconstruction be six (6) months after the final rules are
promulgated. The commenter was concerned that if the final rule
is much less stringent than the proposed rule, operators would
unnecessarily incur significant expenses to install controls that
comply with the proposed rule.
Response; Sections 63.760(f)(2) and 63.1270(d)(2) of the
final rules specify that the compliance date for new and
reconstructed sources (i.e., sources that commence construction
or reconstruction on or after February 6, 1998) is the date of
initial startup, or the effective date of the rule, whichever is
later. Therefore, the EPA believes that new or reconstructed
sources have adequate time to comply with the final rule.
Furthermore, since the final rule is not significantly less
2-219
-------
stringent than the proposed rule (i.e., 95 percent control is
required in the final rule), the EPA believes that owners or
operators should not unnecessarily incur significant expenses to
install controls that comply with the proposed rule.
2-220
-------
2.14 WORDING OF REGULATIONS (OTHER THAN APPLICABILITY AND
DEFINITIONS)
Comment: Commenter IV-D-16 recommended that the EPA delete
the phrase "as expeditiously as practical" in proposed
§63.760(f)(1) as it "is open to interpretation and disagreement
by all." As an alternative, the commenter recommended that
compliance should be by three years after the date of final
rulemaking in the Federal Register. The commenter noted that
this had been done by other MACT standards.
Response; The EPA intends for existing sources to achieve
compliance no fewer than three years after the final rule is
published. It is a benefit to the environment for sources to
achieve compliance in less than three years. However, the EPA
understands that the commenter is concerned about enforcement
actions based on the phrase "as expeditiously as practical." The
EPA does not feel that this phrase adds anything to subparts HH
and HHH and has removed it from §§63.760(f) (1) and 63.1270(d)(1)
of the final rules.
Comment: Commenter IV-D-06 requested that the EPA clarify
the difference between "initial startup" and "startup."
According to the commenter, these two terms do not have the same
meaning. The commenter explained that the time between the
completion of construction and the day of initial startup is used
to try out and debug the equipment. The commenter stated that
these trial runs are not initial startups and should not trigger
the compliance date. The commenter suggested the following
language for §63.1270(d)(2) to clarify and stated that any other
sections in the rule that say "startup" should be changed as
well.
(2) The owner or operator of an affected source
the construction or reconstruction of which commences
on or after February 6, 1998, shall achieve compliance
with the provisions of this subpart immediately upon
initial startup or [the date of publication of the
final rule], whichever date is later.
2-221
-------
The commenter stated that this comment may also apply to subpart
HH.
Response; The EPA agrees that only initial startups should
trigger the compliance dates. Sections 63.760(f)(2) and
63. 1270 (d) (2) of the final rules have been modified to include
the term "initial startup." Additionally, a definition for
initial startup has also been added to §§63.761 and 63.1271 of
the final rules, as follows:
Initial startup means the first time a new or
reconstructed source begins production. For the
purposes of this subpart, initial startup does not
include subsequent startups (as defined in this
section) of equipment, for example, following
malfunctions or shutdowns.
Comment : Commenter IV-D-06 stated that subpart HHH is a HAP
rule and not a VOC rule. The commenter recommended that
references to VOC in the rule should be corrected. The commenter
presented the following changes to §63.1271, and requested that
any other paragraphs with references to VOC also be changed:
Combustion device means an individual unit of
equipment, such as a flare, incinerator, process
heater, or boiler, used for the combustion of volatile
The commenter stated that this comment may also apply to subpart
HH.
Commenter IV-D-07 questioned whether the EPA was suggesting
a need to look at either TOC or HAP, or both TOC and HAP.
According to the commenter, before §63.1281, only HAP is referred
to, while §63.1281 refers to a 95-percent reduction of either TOC
or total HAP.
Response ; In response to commenter IV-D-06, and to be
consistent with the definition of combustion device in other
rules, the EPA has made the following change to §§63.761 and
63.1271.
2-222
-------
Combustion device means an individual unit of
equipment/ such as a flare, incinerator/ process
heater/ or boiler/ used for the combustion of volatile
organic hazardous air pollutant vapors.
The EPA does not intend to regulate VOC or TOO under subparts HH
and HHH, however/ TOC has been included as a surrogate for HAP in
determining control device efficiency. The EPA believes that
allowing the owner/operator to measure TOC rather than HAP
provides some flexibility for the owners and operators and still
achieves the objective of reducing HAP emissions by the specified
emission reduction.
Comment: Commenter IV-D-06 stated that §63.1281 (c) should
say "that closed-vent system must route all HAJ? gases, vapors and
fumes emitted from the reboiler vent to a control device." The
commenter explained that non-HAP gases and emissions from "the
unit," but not from the reboiler vent (such as fugitive
emissions) should not count.
Response; The EPA agrees that the requirement in
§§63.771(c) and 63.1281(c) should specify that closed-vent
systems should route HAP emissions from an affected source, and
has made the recommended changes to subparts HH and HHH.
However, the EPA does not agree the commenter's recommendation
that §§63.771(c) and 63.1281(c) should specify that the emissions
from the reboiler vent be routed to the control device through a
closed vent system. In subparts HH and HHH the standards require
closed-vent systems for more emissions points than just the
glycol dehydration unit reboiler vent. Therefore, the EPA has
not made this recommended change.
Comment: Commenter IV-D-06 pointed out several
typographical errors.
(1) Remove the word "a" from the phrase "... either TOC or a
total HAP in the exhaust gases," in §63.1281(d)(1)(i)(B).
2-223
-------
(2) In Table 1, "Ethylene glycol* should be "Ethylene glycol"
(3) In Table 1, "p-Xylenea" should be "p-Xylene."
The commenter stated that this comment may also apply to subpart
HH.
Response; The EPA has made the recommended corrections.
*
Comment: Cotnmenter IV-D-06 requested that the word
"practicable" in §63.1283 (c) (4) be changed to the word
"practical." According to the commenter,, courts have interpreted
"practicable" to mean "capable of being drone," with little regard
to the cost or other difficulties. The oommenter was concerned
that no matter how quickly they fixed a leak, it would have been
"practicable" to fix it sooner. The commenter stated that they
doubted the EPA intended to impose such a severe requirement.
The commenter also stated that this comment may also apply to
subpart HH.
Response; The word "practicable" has been used in several
regulations in the past (e.g., subpart H
-------
2. Sometimes there will not be any single manufacturer to
provide instructions.
3. Sometimes the manufacturer's instructions will not be
appropriate for the use of the equipment. The
commenter explained that the manufacturer may have
never considered the intended use, or the environment
in which the component would be placed.
The commenter stated that this comment may also apply to subpart
HH.
Response; The EPA agrees with the commenter. Sections
63.773(d)(5) and 63.1283(d)(5) of the final rules specify that
the minimum or maximum operating parameters should be set based
on: (1) a performance test, supplemented by a control device
design analysis or the control device manufacturer's
recommendation or a combination of both [§§ 63.773(d)(5)(i)(A)
and 63.1283(d)(5)(i)(A)], or (2) the control device design
analysis, supplemented by the control device manufacturer's
recommendation [§§ 63.773(d) (5) (i) (B) and 63.1283(d) (5) (i) (B)].
Comment: Commenter IV-D-07 stated that no graphic was shown
for Figure 1 (Section III.A.2, page 6294).
Response; The figure is in the Federal Register version of
the regulation (63 FR 6294).
Comment: Commenter IV-D-07 stated that the referenced
equations are missing in the following sections:
63.772 (a) (4) (iii) (D) , 63 . 772 (a) (4) (iv) (E) , 63 .1282 (d) (3) (ii) ,
63.1282 (d) (3) (iii) , 63 .1282 (d) (4) (ii) (A) , and
63.1282(d)(4)(iii)(B).
Response; The equations are in the Federal Register version
of the regulation (63 FR 6318 and 6331).
2-225
-------
2.15 GENERAL PROVISIONS
Comment: Commenter IV-D-13 stated that additional
rulemaking or preamble language is needed to clarify the
applicability of the General Provisions (40 CFR part 63,
subpart A) to subpart HH. Commenter IV-D-06 stated that
subpart HHH should not defer to subpart A, because according to
the commenter, the General Provisions have significant flaws and
should not be used as the basis for compliance in the MACT
standard. The commenter provided four examples of flaws in the
General Provisions.
1. Section 63.6 (e) (3) (i) (A) says that the startup,
shutdown, and malfunction plan must ensure that sources
are operated in a manner that will minimize emissions
"at least to the levels required by all relevant
standards." The commenter remarked that this provision
could be interpreted two ways, and neither one will
work. According to the commenter, one interpretation
could mean that emissions have to be minimized "as much
as the relevant standards require during startup,
shutdown, and malfunctions." The commenter stated that
this will not work because the relevant standards refer
to §63.6 for the requirements, resulting in an endless
loop, with neither standard stating the requirements.
The second interpretation could mean that emissions
have to be minimized "as much as the relevant standards
require during normal operation when there is not any
startup, shutdown, or malfunction." The commenter
noted that this is impossible, and would eliminate the
reason for having special provisions for startups,
shutdowns, and malfunctions.
2. The general provisions do not specifically address
shutdowns of compliance equipment such as control
devices. According to the commenter, under a literal
reading of a MACT standard, the owner or operator might
simply elect to shut down all the control devices and
assert that no further emission standards apply because
this is a "shutdown."
3. The General Provisions do not specifically address that
some startups, shutdowns, and malfunctions affect only
a portion of the process. According to the commenter,
an owner or operator might assert that a malfunction in
one small, localized portion of a process justifies
shutting down the controls throughout the entire
process, even though the malfunction does not impair
the ability of other portions of the process to comply
with the emission standards.
2-226
-------
4. The General Provisions do not say how to deal with
periods of non-operation when the relevant emissions
have ceased. According to the commenter, a shutdown is
a transitional state between operation and
non-operation. The cotnmenter stated that once the
shutdown is complete, the process is idle and is not
producing emissions. The commenter contended that it
doesn't make sense to impose control requirements when
there are no emissions. The commenter further
explained that some inspectors have interpreted
standards such that control devices must be monitored
even when there is nothing for them to control. The
commenter stated that the HON and other rules clarify
that when there is nothing to monitor (i.e., when the
process is not operating and there are no emissions),
parameter monitoring of control devices is not required
and sometimes a failure to monitor may be excused
during startups, shutdowns, and malfunctions. The
commenter explained that there are instances where
imposing a requirement to continue monitoring would be
inappropriate. For example, monitoring cannot be
continued if the monitoring device has a malfunction or
it might be necessary to "valve off" the monitoring
device to keep the device from being damaged.
Because of the problems with the General Provisions, the
commenter recommended that the EPA should not use the General
Provisions as a basis for handling startups, shutdowns, and
malfunctions. The commenter suggested that the EPA put
provisions equivalent to the HON into §63.1272, which is
currently reserved. The commenter provided the following
language for the new §63.1272:
(a) The provisions set forth and in this subpart
shall apply at all times except during startups or
shutdowns, during malfunctions, and during periods of
non-operation of the affected sources (or specific
portion thereof) resulting in cessation of the
emissions to which this subpart applies. However, if a
startup, shutdown, malfunction, or period of
non-operation of one portion of an affected source does
not affect the ability of a particular emission point
to comply with the specific provisions to which it is
subject, then that emission point shall still be
required to comply with the applicable provisions of
this subpart during the startup, shutdown, malfunction,
or period of non-operation.
(b) The owner or operator shall not shut down
items of equipment that are required or utilized for
compliance with the provisions of this subpart during
2-227
-------
times when emissions are being routed to such items of
equipment, if the shutdown would contravene
requirements of this subpart applicable to such items
of equipment. This paragraph does not apply if the
item of equipment is malfunctioning, or if the owner or
operator must shut down the equipment to avoid damage
due to a contemporaneous startup, shutdown, or
malfunction of the affected source or portion thereof.
(c) During startups, shutdowns, and malfunctions
when the requirements of this subpart do no apply
pursuant to paragraphs (a) and (b) of this section, the
owner or operator shall implement, to the extent
reasonably available, measures to prevent or minimize
excess emissions to the extent practical. For purposes
of this paragraph, the term "excess emissions" means
emissions in excess of those that would have occurred
if there were no startup, shutdown, or malfunction, and
the owner or operator complied with the relevant
provisions of this subpart. The measures to be taken
shall be identified in the applicable startup,
shutdown, and malfunction plan, and may include, but
are not limited to, air pollution control technologies,
recovery technologies, work practices, pollution
prevention, monitoring, and/or changes in the manner of
operation of the source. Back-up control devices are
not required, but may be used if available.
The commenter also suggested adding the words "Except as
otherwise provided in this subpart" to start §63.1281(d)(2). The
commenter stated that this comment may also apply to subpart HH.
Additionally, the commenter requested that the EPA modify Table 2
to list paragraph-by-paragraph the applicability of §63.6(e) to
subpart HHH. The commenter stated that this comment also applies
to subpart HH.
Response; The EPA disagrees with, the commenters that there
are significant "flaws" in the existing subpart A General
Provisions. However, the General Provisions are designed to be
general in nature, and individual NESHAP may have reasons to
override them to implement the intent of the General Provisions
in a standard-specific setting. Consequently, the EPA has
considered the commenters1 concerns related to the startup,
shutdown, and malfunction (SSM) provisions in §63.6(e) and has
added the language suggested by commenter IV-D-06 to §§63.762 and
2-228
-------
63.1272 of the final rules. In addition, §§63.762 and 63.1272 of
the final rule specify that an owner or operator must prepare a
startup, shutdown, or malfunction plan as required in
§63.6(e) (3), except that the plan is not required to be
incorporated by reference into the title V permit.
The commenter is concerned that the requirement for the SSK
plan to ensure that sources are operated in a way that will
minimize emissions "at least to the levels required by all
relevant standards" may result in unclear requirements or will be
impossibly stringent by requiring sources to meet the NESHAP
requirements during all SSM events. The intent of the
requirements in §63.6(e) is that sources do their best to
minimize emissions to the levels of the required standards (i.e.,
the individual subpart). This does not mean, however, that the
source would be required to operate better than the standards or
even to meet the standards during the SSM period, if the source
is in compliance with the SSM plan. Because no plan can cover
every conceivable situation, the duty of the owner or operator is
to do the best he or she can to minimize emissions during all
events, even those not specifically addressed by the plan, based
on good engineering judgement, expertise, and familiarity with
the equipment, as well as following protocols for similar events
that are in the SSM plan, if any.
Commenter IV-D-06 suggested that the General Provisions
could be interpreted to allow a source to shut down control
devices to "eliminate" emissions or to shut down controls for an
entire process if only a portion of that process has a SSM event.
Section 63.2 of the General Provisions defines shutdown as the
"cessation of operation of an affected source for any purpose."
The EPA believes that the general duty to operate and maintain at
2-229
-------
all times, including periods of SSM, "any affected source,
including associated air pollution control equipment, in a. manner
consistent with good air pollution control practices for
minimizing emissions" [§63.6(e)(1)(i)] precludes the possibility
of sources invoking the commenter's interpretation. In fact,
such actions would be considered a violation of the requirements.
The EPA agrees that there are limited cases where continued
monitoring of air pollution control devices during periods of SSM
serves no useful function. These cases should be addressed in
the source's SSM plan, which the State has the authority to
review and approve. However, to address the commenter's concern,
§§63.773(d)(8) and 63.1283(d)(8) of the final rule specify that
emissions during SSM events when the facility is operated in
accordance with the SSM plan, and periods of non-operation of the
unit or process that is vented to the control device (which
result in cessation of emissions), are not considered excursions.
Although the HON requirements do not specifically require
monitoring during periods of SSM, EPA policy is that,
unpromulgated rules should incorporate requirements for
monitoring to be continued at all times. This includes
monitoring during periods of SSM, since these records will
provide the Agency valuable information on the adequacy of the
source's SSM plan and also adherence to the requirement that
emissions are minimized.
Comment; Commenter IV-D-06 requested that the EPA revise
Table 2 to say that §63.7(c) does not apply to subpart HHH.
According to the commenter, the requirement in §63.7(c) of the
General Provisions for a site-specific test plan is "unduly
burdensome" and has not been required in various other MACT
standards. The commenter stated that if subpart HH also
2-230
-------
incorporates §63.7(c), then this comment also applies to subpart
HH.
Response; Site-specific test plans must be developed, but
only need to be submitted to the Administrator for approval upon
request of the Administrator. Subparts HH and HHH do not
specifically require a site specific test plan. Therefore, to
provide State and local permitting authorities the authority to
request site-specific test plans, the EPA has not changed the
applicability of §63.7(c) to subparts HH and HHH.
Comment: Commenter IV-D-16 stated that the requirements of
§63.6 (e) (3) to prepare and follow a startup, shutdown, and
malfunction plan should not be applicable to sites in subpart HH,
since they are likely to be unmanned. The commenter did suggest
that manned gas plants could be required to develop a plan, but
other sources subject to subpart HH should not be subject to
§63.6(e) (3) .
Response; The EPA disagrees that unmanned sites should not
have SSM plans. Such plans are perhaps even more essential at
unmanned sites to ensure that emissions are minimized and
problems addressed as soon as possible. Plans at these sites may
include alarm systems and computerized protocols and other
measures to ensure that appropriate steps are taken to notify
operators of problems and to initiate steps to minimize
emissions. Automated process shutdowns may be required when
certain events occur.
2-231
-------
2.16 MISCELLANEOUS
2.16.1 Health Effects
Comment; Commenter IV-D-28 stated that the EPA should
qualify its statement regarding ethyl benzene's potential
inhalation effects, and reference the many assumptions and
uncertainty factors required in extrapolating from experimental
results in animal studies to potential human effects at actual
concentrations found in the environment.
Response: The proposed rule is technology-based rather than
risk-based, so the summary of toxic effects for ethyl benzene has
no influence on the proposed NESHAP. The EPA included health
effects summaries for ethyl benzene and the other hazardous air
pollutants emitted by this source category to provide the public
with background information about possible effects of
overexposure. The EPA agrees that there are substantial
uncertainties associated with interspecies extrapolation, and has
established national risk assessment guidelines to incorporate
these uncertainties into risk assessments. However, no risk
assessment has been performed in support of the proposed rule.
Thus, the requested explanation of how these uncertainties are
considered, which would necessarily be detailed and long, would
serve no useful purpose.
Comment: Commenter IV-D-28 stated that while the EPA does
not identify the studies on which the conclusions on ethyl
benzene inhalation effects are based, it appears to be relying
principally on adverse effects observed in animal studies.
Response; Given the brevity of the summary, the EPA
believes it has clearly and fairly represented where animal and
human studies have been considered. The use of animal data in
this context is not unusual. Lack of adequate human data for
long-term exposures often makes it necessary to rely on animal
data to predict potential health effects.
2-232
-------
Comment: Commenter IV-D-28 stated that the EPA has
frequently used qualifying language in describing the health
effects associated with chemicals . . . for purposes of other
NESHAP. The commenter suggested that the EPA should include
similar [qualifying] language in this preamble.
Response; The EPA agrees that the suggested qualifying
text, excerpted from the NESHAP for the printing and publishing
industry, accurately reflects the EPA's current thinking on
hazard identification, as expressed in the Agency's guidance on
risk assessment. However, although the EPA believes that the
lack of such qualifying text in this proposal does not
necessarily mislead the public about the purpose of the proposed
rule, this language has been included in the preamble to the
final rule.
Comment: Commenter IV-D-28 stated that the EPA also should
state that its summary is not intended to be relied on to
characterize ethyl benzene's potential inhalation toxicity.
Response; The EPA has not suggested anywhere in the
proposed rule that this brief paragraph should serve as a risk
characterization. The section presents only a qualitative
description of effects that may result from overexposure, and
does not suggest where, when, or if such overexposure may occur.
Comment: Commenter IV-D-28 stated that the preamble fails
to include any review of the scientific database on ethyl
benzene's toxicological effects, or reference any animal or human
inhalation studies of ethyl benzene. According to the commenter,
absent such a review the EPA should not publish, in a Federal
Register notice, findings regarding a chemical's health risks.
Response; The EPA agrees with the coxmnenters' point that
risk assessment findings should be fully supported by a review of
the toxicological database. However, the 5-sentence summary of
toxic effects cannot be construed as a risk assessment for ethyl
2-233
-------
benzene. It contains only qualitative descriptions of potential
effects of overexposure, and presents no findings regarding risks
from ethyl benzene exposure (or any information at all about
exposure) . This paragraph does not need to be, and makes no
pretense of being, a fully-referenced review of the scientific
database.
Comment: Commenter IV-D-28 stated that the EPA states that
"short-term inhalation of high levels of ethyl benzene in humans
may cause throat and eye irritation, chest constriction, and
dizziness." The commenter requested that the EPA delete this
statement from the preamble to the final rule. The commenter
further stated that if the EPA does not delete the discussion
entirely, the EPA should, at a minimum, acknowledge the
substantial database demonstrating the absence of acute effects
following inhalation .exposure to ethyl benzene.
Response; The EPA based its description of short-term
respiratory and ocular effects on a review of the toxicological
literature by the U.S. Agency for Toxic Substance and Disease
Registry (ATSDR),28 ATSDR's supported its description of throat
irritation, chest constriction, and burning eyes accompanied by
profuse lacrimation at ethyl benzene levels above 1000 parts per
million with numerous literature citations. The commenters1
citation of other studies that failed to show such effects, while
potentially interesting, does not change the results of studies
that did report these effects.
Comment: Commenter IV-D-28 stated that the EPA also should
acknowledge that levels of ethyl benzene in the ambient air are
well below those at which adverse health effects have been
observed.
28 Agency for Toxic Substance and Disease Registry , 1997.
Toxicological Profile for Ethyl benzene. Public Comment Draft.
Toxicology Information Branch, Atlanta, GA
2-234
-------
Response; The EPA's preamble to the proposed rule
(paragraph 1, last sentence) states, "In general, these findings
have only been shown with concentrations higher than those in the
ambient air." This statement applies to ethyl benzene as well as
the other hazardous air pollutants for which the preamble
describes health effects. The EPA believes that this statement
already addresses the commenter's concerns in a manner that is
more than fair. Because the EPA lacks positive evidence that all
locations in the United States are free from hazardous ambient
concentrations of ethyl benzene, it is impossible to further
strengthen this statement.
Comment: Commenter IV-D-28 requested that the EPA delete,
or qualify substantially, its statement regarding ethyl benzene's
teratogenic effects in animals to reflect the toxicological
database accurately.
Response; ATSDR's 1997 review of the health effects of
ethyl benzene,29 describes several animal inhalation studies that
reported delayed skeletal development, increased incidence of
extra ribs, and renal malformation. The EPA believes that the
sentence currently in the preamble, "Birth defects have been
reported in animals exposed via inhalation; whether these effects
may occur in humans in not known," is a fair condensation of
ATSDR's review. As with the acute effects above, the commenter's
citation of a study for which no teratogenic effects were found
does not change the conclusion, based on positive data, that such
effects are possible.
Comment: Commenter IV-D-28 requested that the EPA delete,
or revise substantially, its discussion of ethyl benzene's
potential chronic health effects. According to the commenter, the
29 Reference 28 .
2-235
-------
EPA's statement does not accurately describe ethyl benzene's
potential hematological effects and may potentially mislead
users, consumers, and regulatory bodies about the health risks
associated with repeated exposure to ethyl benzene.
Additionally, the EPA's statement is inconsistent with the
relatively low toxicity assigned to ethyl benzene under the EPA's
Sector Facility Indexing Program (SFIP) .
Response; The EPA agrees that the language in the current
preamble does not describe ethyl benzene's chronic effects well,
or in fact at all. Reported effects to the rodent kidney include
increased kidney weight and increased activity of several kidney
enzymes. Hepatic effects reported in animal studies include
increased liver weight, altered enzyme activity, and degenerative
changes in liver cells. Hematologic effect reported in workers
occupationally exposed to ethyl benzene include decreased
hemoglobin levels and increased lymphocyte count. The EPA
believes the current brief description, "Animal studies have
reported blood, liver, and kidney effects associated with ethyl
benzene inhalation." is consistent with these findings and does
not think additional background detail is needed to support the
proposed rule.
The EPA also agrees that ethyl benzene has relatively low
toxic potential when compared with many other substances listed
as hazardous air pollutants under section 112 of the Clean Air
Act. However, the proposed action is not based on a risk
assessment, and the preamble language therefore does not intend,
or need, to convey information about risk. The one-sentence
listing of chronic health effects refers only to adverse effects
that may occur if the dose of ethyl benzene is high enough. Any
further discussion of relatively low toxicity would need to be
accompanied by information about the large amounts of ethyl
2-236
-------
benzene emitted relative to other, more toxic, pollutants, none
of which would be germane to the proposed rule.
Comment: Commenter IV-D-28 requested that the EPA clarify
that adverse health effects have been observed only at ethyl
benzene levels substantially higher than those found in the
ambient air.
Response; The coxnmenters' data on ethyl benzene
concentrations is generally consistent with the EPA's data. The
EPA agrees that our current understanding, based on these air
monitoring data and on modeled concentrations, suggests that
ethyl benzene does not currently pose a national threat.
However, measured concentration data are sparse, and detailed
modeling has not been conducted for many areas. These very
substantial data gaps prevent the EPA from conclusively stating
that no location in the United States experiences ethyl benzene
levels above the reference concentration.
Furthermore, the National Toxicology Program (NTP) recently
released data that suggest ethyl benzene may be carcinogenic in
animals. If further studies bear this out, perceived "safe"
levels of ethyl benzene may change substantially. This
uncertainty about carcinogenicity also prevents any categorical
guarantee of safety from health effects of ethyl benzene. The
EPA believes the current statement, "In general, these findings
have only been shown with concentrations higher than those in the
ambient air," is the strongest that can accurately be made.
2.16.2 Other Miscellaneous Comments
Comment: Commenter IV-D-06 supported semiannual reporting
of actions inconsistent with the startup, shutdown, and
malfunction plan instead of immediate reporting and suggested
that the EPA apply the concept to all MACT standards (including
standards that have already been promulgated). The commenter
2-237
-------
cited three reasons that immediate reporting has been an
unnecessary burden. First, the commenter contended that just
because actions are inconsistent with the startup, shutdown and
malfunction plan, does not mean that excess emissions have
occurred. Second, even if excess emissions have occurred, they
are not necessarily an immediate health threat (most HAP concerns
are related to long-term, rather than short-term, exposure) .
Third, in any case where short-term emissions may be a concern,
there are better ways to address those concerns (including
immediate reporting under CERCLA section 103 or SARA section 304,
or emergency response actions under EPA's Risk Management
Programs rule implementing section 112(r) of the Act). The
commenter stated that retrofitting semiannual reporting into
existing MACT standards should not be a major burden on the EPA
as it is a minor change and would not likely be controversial.
Response; While the EPA is continually reviewing the
burdens imposed by its regulations in an attempt to reduce these
burdens, no changes have been identified at this time.
Comment: Commenter IV-D-38 stated that the proposed rules
exceed the EPA's authority under section 112 of the CAAA and the
intent of Congress in promulgating the CAA.
Response; The EPA is required to develop NESHAP under
section 112(d) of the CAA for listed source categories and has
not exceeded its authority.
Comment: Commenter IV-D-04 stated that based on the data
presented by the EPA, there are many small natural gas
production, transmission and storage facilities that are small
sources of HAP. According to the commenter, these sources are
already regulated under State rules and the commenter requested
that the EPA not regulate these sources further.
Response; When developing the MACT floor for production and
transmission and storage facilities, the EPA made several
distinctions between sizes of emissions units. For example, the
EPA developed a MACT floor for large glycol dehydration units
2-238
-------
(natural gas flowrate greater than or equal to 3 MMscf/d or
benzene emissions greater than or equal to 1 tpy) and one for
small glycol dehydration units (natural gas flowrate less than 3
MMscf/d or benzene emission less than 1 tpy). In this case, the
MACT floor for small glycol dehydration units was determined to
be no control and these units are not required, in the final
rules, to be controlled. Therefore, the NESHAP are focussed on
facilities with significant HAP emissions. These criteria will
exempt a large number of facilities in the oil and natural gas
production and natural gas transmission and storage source
categories from regulatory requirements and installation of
controls due to their small size and low emissions. These
applicability criteria include benzene emission rate, natural gas
throughput, storage tank throughput, and hours of operation and
apply only to major sources.
2-239
-------
2.17 GENERAL COMMENTS SPECIFIC TO SUBPART HHH (NOT OTHERWISE
ADDRESSED)
Comment: Several commenters were concerned with the EPA's
proposed standard for the natural gas transmission and storage
source category and the EPA's apparent lack of. information used
to develop the standard. Commenter IV-D-31 stated that the
inclusion of natural gas transmission and storage in the oil and
natural gas production rulemaking has compromised the regulatory
process and denied affected stakeholders equal opportunity for
input. Although the commenter was pleased that the EPA created a
separate source category for the transmission and storage
industry, they suggested that the EPA did not truly develop
separate MACT standards for the two source categories.
Furthermore, the commenter stated that including the transmission
and storage source category in this proposal is inconsistent with
the procedures of the CAA of 1990. According to the commenter,
the EPA extended the oil and natural gas production source
category, without adequate notice and opportunity to comment to
include the natural gas transmission and storage sources.
The commenter stated that the EPA is obligated to provide
equal opportunity for the natural gas transmission and storage
industry to work with the EPA during the rulemaking process. The
commenter further stated that regulating both sources
simultaneously, to save limited resources, does not justify
"abandoning well-established procedures for developing a MACT
that is achievable." The commenter stressed that in raising
these issues they are not attempting to avoid or delay regulation
but insisting on the right to the benefit of the full MACT
development process, including source category listing, data
gathering, determination of a MACT floor, and source
category-specific standards.
Commenters IV-D-07, IV-D-12, IV-D-31, IV-G-09 stated that
the EPA has insufficient data to develop a standard for the
transmission and storage source category. The commenters
requested that the EPA delay the transmission and storage portion
of the rulemaking to properly survey the industry for more
2-240
-------
meaningful data, and assess whether a standard for the natural
gas transmission and storage source category is necessary or
achievable. According to commenter IV-D-07, the EPA would create
confusion "by leaving numerous questions unanswered and terms
undefined or interrelated with the proposed oil and gas
production standard" in its haste to develop a standard for the
natural gas transmission and storage source category
simultaneously with that of the production source category.
Additionally, commenters IV-D-07 and IV-D-31 stated that the EPA
has proven a lack of information on this source category by
asking for emission point comments on reboiler vents, flash tanks
(GCG separators), storage vessels with flash potential, pipeline
pigging and storage of pipeline pigging wastes, and equipment
leaks.
Response; The EPA contacted potential stakeholders in the
initial phase of the development process for this NESHAP to
identify a list of interested stakeholders. The public record,
contained in EPA Air Docket A-94-04, has correspondence and
meeting summaries that show that the EPA had continual contact
with interested stakeholders, including representatives of the
natural gas transmission and storage source category. However,
to address industry concerns on the adequacy of the database used
in the development of proposed subpart HHH for natural gas
transmission and storage facilities, the EPA has collected
additional information on glycol dehydration units in the natural
gas transmission and storage source category. The EPA conducted
site visits to five natural gas transmission and storage
facilities to gain additional first-hand knowledge of the
processes and operations at existing facilities in this source
category. The EPA also met with stakeholders from the natural
gas transmission and storage industry to understand their
concerns. The EPA developed a questionnaire for distribution to
2-241
-------
eight natural gas transmission and storage companies under the
authority of section 114 of the CAA. In the questionnaire, the
EPA requested data on the processes, operations, and control
technologies in use at existing natural gas transmission and
storage facilities and relevant to the development of HAP
emissions standards for glycol dehydration units.
Through the questionnaire and site visits, the EPA collected
additional information on approximately 83 glycol dehydration
units in the natural gas transmission and storage source
category. The EPA considered this new information, along with
the previously collected information on the natural gas
transmission and storage source category, to develop a MACT floor
for process vents on glycol dehydration units located at existing
and new facilities in this source category.
On January 15, 1999, the EPA published a supplemental notice
announcing the availability of and to discuss the consideration
of, the additional information on the natural gas transmission
and storage source category collected by the EPA since proposal
(64 FR 2612). The additional data announced in the January 15,
supplemental notice included the following items located in Air
Docket A-94-04: (1) completed responses to the EPA's section 114
survey questionnaire, items IV-G-24, and IV-G-26 through IV-G-32;
(2) site visit information, items IV-G-21, IV-G-22, and IV-G-25;
and (3) meeting summary of the meeting with representatives of
the Interstate Natural Gas Association of America, the Gas
Research Institute, and industry, item IV-E-02. The EPA has also
prepared analyses of these data, items IV-A-01, IV-A-02 and
IV-A-08.
Comment; Commenter IV-D-31 stated that the EPA
underestimated the impact of the proposed regulation on the
2-242
-------
natural gas transmission and storage source category, due to the
EPA's lack of sufficient and representative information to
develop a MACT standard. The commenter contended that the EPA
underestimated the number of transmission and storage facilities
that would be affected by subpart HHH (five out of 2,000)
resulting in an underestimation of the cost impact. The
commenter suggested that the EPA postpone the natural gas
transmission and storage NESHAP (subpart HHH) to evaluate the
industry properly and to develop a better estimate of what the
MACT floor should be. The commenter offered to work with the EPA
on the development of a separate proposal.
Response; As stated in the previous response, in response
to comments on proposed subpart HHH, the EPA surveyed eight
natural gas transmission and storage companies under the
authority of section 114, conducted five site visits to
transmission and storage facilities, and received additional
information on 83 glycol dehydration units. In addition, the EPA
had data on 31 glycol dehydration units that was collected during
the development of the proposed NESHAP. According to the Oil and
Gas Journal. 3° the total natural gas throughput handled by the
companies for which the Agency had information represented
approximately 14 percent of the total natural throughput for the
entire industry. Of the 114 glycol dehydration units for which
information was submitted, the EPA determined that one unit had
the potential to be an affected source under subpart HHH.
Based on this information, the EPA projected the number of
affected sources to a nationwide value. Since the available data
represented approximately one seventh of the industry, the EPA
estimated that seven glycol dehydration units would be affected
30 True, W.R. Weather, Construction Inflation Could Squeeze
North American Pipelines. Oil & Gas Journal, pp. 33-56. August
31, 1998.
2-243
-------
sources. Although the MACT floor indicated that the best
performing 12 percent of the existing sources used combustion
(primarily flares) to control HAP emissions from the glycol
dehydration units, the EPA assumed that at least one of the
affected facilities would install a condensation unit to control
HAP emissions. Thus, the environmental and cost impacts were
based on six of the facilities installing a flare to meet the
requirements of subpart HHH and one would install a condenser.
Table 2.17-1 presents a summary of estimated environmental,
energy, and cost impacts for the natural gas transmission and
storage standards for existing major sources. The impacts were
revised using the same approach that was used for the proposed
NESHAF. A detailed analysis regarding the estimated impacts of
2-244
-------
TABLE 2.17-1
SUMMARY OF REVISED ESTIMATED ENVIRONMENTAL, ENERGY, AND COST
IMPACTS FOR THE NATURAL GAS TRANSMISSION AND STORAGE STANDARDS
FOR EXISTING MAJOR SOURCES*
Existing Natural Gas
Transmission and
Impact category Storage*
-Estimated—number of impacted 3—
facilities
Emission reductions (Mg/yr)
HAP 390
VOC 610
Methane 230
Secondary environmental emission
increases (Mg/yr)
Sulfur oxides <1
Nitrogen oxides <1
Carbon monoxide <1
Energy (Kilowatt hours per year) None
Implementation costs (Million of July 1993 $)
Total installed capital 0.28
Total annual 0.3
* - No new major sources are anticipated for this source
category after the effective date for new sources and in the
first three years following promulgation of the rule.
2-245
-------
the NESHAP is presented in the docket (Air Docket A-94-04, number
IV-A-08).
Comment: Commenter IV-D-31 was concerned that the EPA is
regulating the natural gas transmission and storage industry
similarly to the oil and natural gas production industry.
According to the commenter, the EPA's "skewed data collection
effort" resulted in a failure to make important distinctions
between the transmission and storage and the production segments
of the industry. The commenter was especially concerned about
the potential impact of subpart HHH on underground storage
facilities, as they are used to offset fluctuations in gas flow,
reduce natural gas costs, and improve reliability. Commenters
IV-D-07, IV-D-12, and IV-D-31 explained that a review of the
background information for this standard showed that the database
consisted of information on the methods used in natural gas
transmission from only two companies and no underground storage
facilities. The commenters noted that the companies surveyed
were oil production facilities that handled gas as a by-product
of oil production that have higher HAP emissions because they
handle more liquids with higher concentrations of HAP.
Furthermore, commenters IV-D-31 and IV-G-12 also noted that once
the gas reaches the transmission and storage facilities, it has
been dehydrated at least once, further lowering the
concentrations of HAP. Commenter IV-G-12 also mentioned that
exposing processed gas to ground water in a storage facility can
increase the moisture content and require additional dehydration,
but it does not necessary increase BTEX in the gas. The
commenter referred to the GLYCalc manual for discussion of impact
of BTEX concentration in the gas on dehydrator emissions.
Commenters IV-D-10 and IV-G-12 recommended changes in the
exemptions for transmission and storage facilities. Commenter
IV-D-10 requested that the size cutoff for transmission and
storage be higher than that for production, since HAP emissions
at transmission and storage facilities are generally much lower
than production facilities. Commenter IV-G-12 recommended that
2-246
-------
the subpart HHH provide an exemption for transmission and storage
facilities where the BTEX concentrations of the stored gas fall
below a minimum threshold and it can be shown that the act of
storing the gas does not cause an increase in the concentration
of the BTEX in the gas when it is retrieved from storage. The
commenter further explained that dehydrators serving transmission
and storage facilities are fundamentally different from those
located at production wells.
Response; In the proposal, glycol dehydration units
operating at an actual annual average natural gas throughput less
than 3 MMscf/d or having actual average benzene emissions less
than 1 tpy were exempt from the control requirements. The EPA
evaluated the data collected for 114 glycol dehydration units in
the natural gas transmission and storage source category to
determine whether there was a natural gas throughput level, or a
benzene emission level for which glycol dehydration units
operating below this level were not controlled.
In the new data, the Agency did not identify evidence to
suggest that glycol dehydration units operating with actual
annual average natural gas throughput rates less than 10 MMscf/d
or having actual benzene emissions less than 1 tpy are controlled
at the MACT floor and it was not cost effective to go beyond the
floor for these glycol dehydration units.
In addition, the Agency does not have any information
indicating that there are any sources in the natural gas
transmission and storage source category operating below
10 MMscf/d or having benzene emissions less than 1 tpy that have
emissions greater than the major source thresholds of 10 tpy for
individual HAP or 25 tpy for any combination of HAP.
Therefore, the final subpart HHH exempts each glycol
dehydration unit with an annual actual average natural gas
2-247
-------
throughput less than 10 MMscf/d or actual average benzene
emissions less than 1 tpy from the control requirements.
Comment: Commenter IV-D-31 stated that they believe that
the EPA lacks sufficient and representative information on the
natural gas transmission and storage source category to determine
a MACT floor. The commenter noted that only one transmission
facility, and no underground storage facilities, was surveyed.
The commenter recommended that the EPA revise its MACT floor
determination for transmission and storage dehydration units.
The commenter suggested that the EPA did not include the
additional cost that would be required to send personnel to
remote, unmanned transmission and storage facilities to
demonstrate compliance based on short averaging periods. The
commenter stated that a longer averaging period would also be
more practical and cost-effective, and more likely to be
achievable by the best performing sources. The commenter asked
the EPA to take the time to gather the information required to
properly evaluate what is actually achieved by emission points in
the natural gas transmission and storage source category.
Response: According to the information collected from 114
glycol dehydration units through the section 114 questionnaire,
site visits, and data previously collected during the development
of the proposed standards, 71 glycol dehydration units are
controlled. Fifty-nine of these units utilize combustion as the
control technology for process vents on glycol dehydration units.
Of these, 51 utilize flares, seven utilize enclosed combustion
devices, and one uses an in-stack flare system. Seven units
utilize a combination of condensation and combustion to control
glycol dehydration unit process vents and five utilize
condensation.
The MACT floor analysis for the natural gas transmission and
storage source category was based on information available on the
2-248
-------
top 14 performing glycol dehydration units/ which corresponds to
12 percent of 114 glycol dehydration units.
The EPA compared the control level data for the top 14
performing units to the proposed control level of 95 percent for
process vents on glycol dehydration units at existing and new
natural gas transmission and storage facilities. The available
information indicates that the best performing 12 percent of the
facilities, i.e., 14 units, utilize some form of combustion and
achieve a HAP emission reduction of at least 98 percent.
However, among all sources that apply combustion, the reported
control efficiency ranged from 95 to 98 percent. The EPA was
unable to determine the technical basis for the reported
differences in the control efficiencies for these combustion
devices. Therefore, in order to account for the observed
variability in HAP emission reduction efficiency, the final rule
requires 95 percent as the HAP emission reduction for this source
category associated with this technology.
Under the proposed standards, the MACT floor for new sources
was the same as the MACT floor for existing sources (i.e., 95-
percent control). In the review of the new additional
information, the EPA did not identify a method of control
applicable to all types of new sources that would achieve a
greater level of HAP emission reduction than the MACT floor for
existing sources. Therefore, as with the proposal, the EPA
determined that the MACT floor for new sources in the natural gas
transmission and storage source category was the same as the MACT
floor for existing sources.
Comment.- Commenter IV-D-31 supported the aggregation of
equipment at compressor stations and single wells, with their
associated equipment, for major source determinations. However,
2-249
-------
the commenter explained that natural gas storage fields cover
large areas and operations are based on depleted production
fields with the same surface separation of facilities as existing
production fields. Therefore, the commenter, along with commenter
IV-D-04, stated that aggregating emission sources at natural gas
storage facilities for major source determinations is
inconsistent with the meaning of section 112 (n) (4) (A) of the CAA
and that aggregating emission sources at these facilities
suggests that the EPA has insufficient data regarding natural gas
storage facilities. The commenters explained that glycol
dehydrators at storage facilities do not emit significant
quantities of HAP because the natural gas has been processed or
dehydrated before it is injected into the storage fields, and has
small amounts of HAP and VOC. According to the commenters,
aggregating affected sources over large gas storage fields could
result in major source determinations and the controls on each
contributing affected source would be more expensive than the EPA
has estimated, for small amounts of HAP reduction. The commenters
recommended that the EPA collect data to characterize natural gas
storage operations better.
Response; Section 112(n)(4)(A) of the Act states that
". . .emissions from any oil or gas exploration or
production well (with its associated equipment) and
emissions from any pipeline compressor or pump station
shall not be aggregated with emissions from other
similar units . . . ."
The EPA has interpreted this provision to mean that individual
pipeline compressor or pump stations shall not be aggregated with
emissions from other stations. Nothing in the
section 112(n)(4)(A) provisions refers to natural gas storage
facilities as those facilities for which emission aggregation is
not allowed. Additionally, the definition of major source in
§63.1271 states that emissions from processes, operations, or
equipment that are not part of the same facility shall not be
2-250
-------
aggregated. Based on the EPA's knowledge of the industry,
storage fields should have well-defined surface sites, preventing
large areas from being considered part of the same facility.
Cpmment: Commenters IV-D-10 and IV-D-31 requested that the
deadline for promulgation of the transmission and storage
standard be changed from November 15, 1997 to November 15, 2000.
The commenters noted that transmission and storage is a new
source category.
Response; The EPA amended the source category list to add
the natural gas transmission and storage source category as a
major source category and proposed a regulation that would apply
to major sources in this source category. As stated in section
112(c)(5) of the Clean Air Act as Amended in 1990 on the addition
of source categories
". . .emission standards under subsection (d) for the
category or subcategory shall be shall be promulgated
within 10 years after enactment of the Clean Air Act
Amendments of 1990, or within 2 years after the date on
which such category or subcategory is listed, whichever
is later."
Although the natural gas transmission and storage source category
is in the 10-year bin of source categories, by promulgating the
proposed regulation on the current schedule, the EPA is complying
with the requirements of the CAA.
Comment; Commenters IV-D-07 and IV-D-31 stated that the
definition of natural gas transmission is misleading and mixes
production terms with natural gas transmission terms. For
example, "boosters" are only used on production lines. The
commenters further remarked they did not understand the phrase
"used for long distance transport" since "long distance" is not
defined. The commenters recommended that the EPA use the
Department of Transportation's definitions (49 CFR 192.3) for
such terms as pipeline, transmission line, and transportation of
natural gas.
2-251
-------
Response; The EPA does not believe that further
clarification of the definition for natural gas transmission is
necessary. The definition of natural gas transmission in
§63.1271 of subpart HHH was developed in consultation with
stakeholders in the natural gas transmission and storage source
category (Air Docket A-94-04 Numbers II-C-4, II-C-5, and
II-D-53) . As requested by the stakeholders, this definition is
consistent with the definition used by the Federal Energy
Regulatory Commission and adequately reflects the actual workings
of the industry*
In addition, the EPA's understanding is that natural gas
transmission pipelines differ from natural gas mains in that titey
typically operate at higher pressure, are longer, and the
distance between connections is greater. Therefore, the EPA has
retained the phrase "long distance transport" to maintain this
distinction.
Comment: Commenter IV-D-16 stated that the term associated
equipment does not need to be defined in subpart HHH since
production sources are not covered. Commenter IV-D-35 agreed with
proposed definition of associated equipment.
Response; The EPA agrees that the term associated equipment
is unnecessary and has removed this term from §63.1271 of subpart
HHH. In addition, the EPA has revised the definition of major-
source in §63.1271 as follows:
Major source, as used in this subpart, shall have
the same meaning as in §63.2, except that:
(1) Ernieeiono from any oil eg gao exploration or
production well (with its associated equipment)
eademissions from any pipeline compressor or pump
station shall not be aggregated with emissions from
other similar stationsunitea. whether or not such, units
are in a contiguous area or under common control; and
2-252
-------
(2) emissions from processes, operations, and
equipment that are not part of the same facility, as
defined in this section, shall not be aggregated.
Comment: Commenter IV-D-16 stated that a compressor station
is defined in §63.1271 as equipment that "...supplies energy to
move natural gas at increased pressure from fields. ..." The
commenter requested that the reference to fields be deleted since
subpart HHH does not cover production sites (i.e., fields).
Response; As suggested by the commenter, subpart HHH does
not encompass production sites. However, it is possible for
compressors, located in the transmission and storage source
category, to move natural gas from production fields.
Furthermore, the EPA developed this definition based on standard
industry nomenclature. Therefore, the EPA does not believe that
the change recommended by the commenter is warranted.
2-253
-------
2.18 COMMENTS RECEIVED ON THE JANUARY 15, 1999 SUPPLEMENTAL
NOTICE (64 FR 2611)
Comment: Commenters IV-G-38 and IV-G-37 indicated that they
did not agree with a MACT floor of 95 percent for the
transmission and storage source category. The cotnmenters stated
that capital expenditures had been made to control some existing
dehydrators with condensers and that it would be unreasonable to
require these sources to meet a MACT floor based on a different
technology.
The commenters requested that the final rule should either
exempt existing sources controlled by condensers, or require that
existing sources controlled with condensers be controlled to a
different level (i.e., 70 percent) than the combustion
technology-based MACT floor. The commenters stated that the data
show that condensers could consistently achieve a 75-percent
emission reduction and that requiring an additional 20-percentage
points of emission reduction in HAP would be inconsistent with
the cost-to-benefit analysis in the February 6, 1998 proposal.
In addition, commenter IV-G-36 stated that the 95 percent control
level cannot be continuously achieved by the use of condensers
alone.
Response; The EPA does not believe that it is appropriate
to provide exemptions or alternative levels of control for
existing glycol dehydration units that are controlled by
condensers. The EPA believes that this would not be consistent
with the Act, which specifies in section 112(d)(3) that for a
source category with 30 or more sources (such as the transmission
and storage source category), the MACT floor for existing sources
shall not be less stringent than the "... the average
limitation achieved by the best performing 12 percent of the
existing sources. ..." The data collected by the EPA indicated
that the average limitation achieved by the top 12 percent of the
existing glycol dehydration units located at natural gas
2-254
-------
transmission and storage facilities was 95 percent. Furthermore,
the data indicated that the top 12 percent of the existing glycol
dehydration units were controlled using combustion or a
combination of combustion and condensation. Therefore, in
accordance with the statute, the EPA established the MACT floor
to be 95 percent for glycol dehydration units located at natural
gas transmission and storage facilities, which corresponds to
combustion.
With regard to the comment regarding continuous compliance,
in the supplemental notice the EPA did not address the issue of
averaging period for condensers in use at transmission and
storage facilities. The final subpart HHH allows an owner or
operator that installs a condenser for control of HAP from glycol
dehydration unit process vents to establish compliance with the
95-percent HAP emission reduction on a 30-day rolling average.
In addition, the final subpart HHH allows the owner or operator
to comply with: (1) 95 percent HAP emission reduction,
(2) 20 ppmv outlet HAP concentration for combustion devices, or
(3) outlet emissions of 1 tpy of benzene. The EPA believes that
the 1 tpy benzene emission limit and the 30-day averaging period
for condensers provides sufficient flexibility for owners and
operators of existing controlled glycol dehydration units.
Comment: Comrnenters IV-G-37 and IV-G-38 referred to the
proposed rule (63 FR 6288), which provided a 20 ppmv outlet HAP
concentration limit for combustion devices. The commenters
stated that the EPA did not provide rationale for dropping this
limitation and requested that it be retained.
Response; The EPA did not drop the 20 ppmv requirement for
combustion devices. The final rule requires owners or operators
to meet: (1) a 95-percent HAP emission reduction, (2) a 20 ppmv
2-255
-------
outlet HAP concentration limit for combustion devices, or (3) a
1 tpy outlet benzene emission limit.
Comment: Commenters IV-G-37, IV-G-38 ,and IV-G-39 agreed
that exempting glycol dehydration units with actual annual
average natural gas throughputs less than 283 thousand m3/day
from the control requirements under subpart HHH was appropriate.
Commenters IV-G-37 and IV-G-38 stated that they were unaware of
any dehydration units that operate at the higher flow rate that
would exceed the HAP emission cut-off. In contrast, commenter
IV-G-36 stated that a cutoff level of 10 MMscf/d would provide no
significant relief to the majority of companies in the
transmission and storage segment of the industry. Although they
supported the change in the level for the cutoff, the commenter
stated that dehydration units of this size would probably be
exempt from the controls based on the criteria for major sources.
Response; The EPA appreciates the Commenters1 support and
the EPA believes that the cutoff level of 10 MHscf/d, which is
based on the MACT floor determination, is appropriate for this
industry. However, commenter IV-G-36 was incorrect in stating
that dehydration units with natural gas throughput less than 10
MMscf/d would be exempt from controls based on the criteria for
major source. This statement is true if the glycol dehydration
unit is the only HAP emission source located at the facility.
However, if the facility is determined to be major source due to
the aggregation of all HAP emission sources located at the
facility, then the owner or operator must comply with the control
requirements for each glycol dehydration unit at that facility.
Thus, only glycol dehydration units that have natural gas
throughput less than 10 MMscf/d are exempt from control
requirements at the major source.
Commenter; Commenter IV-G-36 referred to Section I
(Background) of the supplemental notice, which indicated that the
2-256
-------
original data included a questionnaire to one company with 31
glycol dehydration units. The commenter stated that they had
assumed that this corresponds to the September 15, 1993 entry
noted in table A-l of the BID, Questionnaire to CNG Transmission
Corporation.
Response; The commenter's assumption is correct. The EPA
surveyed CNG Transmission Corporation and received information
for 31 glycol dehydration units.
Comment; Commenter IV-G-36 reviewed the docket items listed
in the supplemental notice as containing the information used by
the Agency to evaluate the transmission and storage source
category (Air Docket A-94-04 Numbers IV-E-02, IV-G-21, IV-G-22,
and IV-G-24 through 32). The information contained in item
IV-G-26 was deemed to be confidential. The commenter stated that
the data they were able to review indicated that there were 89
additional glycol dehydration units, 14 of which were noted as
having emission controls. The commenter stated that it was
unclear how the EPA derived the numbers presented in Section III
(MACT Floor for Existing Sources) regarding the total number of
units and the number of controlled units. According to the
commenter, the data they were able to review indicated that there
are a minimum of 120 dehydration units (instead of the 112
reported by the EPA) of which 14 have been identified as having
controls (instead of 69 as reported by the EPA). Furthermore,
the commenter indicated that there was a discrepancy in the types
of controls being utilized. However, the commenter also noted
that the data on the original 31 units and the confidential
information contained in docket number IV-G-26 were not reviewed.
Response; The discrepancy in the number of units that the
commenter identified and the number of units reported in the
supplemental notice resulted from the information not reviewed by
the commenter including: the confidential information (i.e.,
information contained in site visit reports, and the information
collected from one company under the authority of section 114),
2-257
-------
the information collected during the development of the proposal,
as well as the number of units for which no information was
submitted as instructed in the section 114 questionnaire.
Table 2.18-1 presents a summary of the information collected to
identify the number of existing glycol dehydration units and the
types of controls in use. As shown in the table, the EPA used
three sources of information: section 114 questionnaire (data
received 12/93), site visits (conducted in 10/98), and section
114 questionnaires (data received 10/98).
The number of glycol dehydration units identified in the
response to the section 114 questionnaire in December 1993 is
contained in the docket (Air Docket A-94-04 number II-D-26) .
The data collected under the October 1998 section 114
questionnaire is also contained in the docket (Air Docket A-94-04
2-258
-------
IT)
CM
I
CN
-------
numbers IV-G-24 and IV-G-26 through 32) . Respondents to the
October 1998 section 114 questionnaire were instructed to submit
all information on glycol dehydration units that have natural gas
throughputs greater than 15 MHscf/d. Respondents were also
instructed that the requested information was not required for
each glycol dehydration unit where the natural gas throughput was
less than 15 MMscf/d, but could be submitted if available. The
respondents identified 30 glycol dehydration units that had
natural gas throughputs less than 15 MMscf/d, and no information
was submitted for these units (it should be noted that
information was provided for three glycol dehydration units for
which the natural gas throughput was less than 15 MMscf/d) .
In addition, while reviewing the number of dehydration units
for which the EPA had information, the EPA identified two
additional glycol dehydration units, both of which were
controlled (Air Docket A-94-04 numbers IV-G-21, IV-G-22, and
IV-G-25). Therefore, the total number of glycol dehydration
units for which the EPA has information is 114 (compared to 112
as stated in the January 15 supplemental notice).
Comment: Commenters IV-G-36, IV-G-37, and IV-G-38 referred
to a GRI report31 which stated that condensers could not
consistently achieve a 95-percent reduction of HAP. The
commenters indicated that this report had been submitted to the
EPA and was included in the regulatory record. Commenters
IV-G-37 and IV-G-38 stated that in the supplemental notice, the
EPA did not present information refuting that condensers may not
31 The commenters referred to the following draft report:
"BTEX Emission from T&S Industry Segment Glycol Dehydrator."
July 27, 1998, by Radian International LLC. Conversations with
the commenters (Air Docket A-94-04 numbers IV-A-10 and IV-A-11)
indicated that the final report title is "Glycol Dehydrator
Emissions when Treating Low-BTEX Gas." November 5, 1998.
2-260
-------
achieve the proposed HAP emission reduction, but had instead
chosen to change the technological basis for the MACT floor for
existing and new sources.
Commenter IV-G-36 was concerned that the EPA did not
consider all of the information when evaluating the proposed
standards. The commenter stated the information, compiled by GRI
regarding emission levels as functions of gas throughput, gas
temperature, gas pressure, water content, and BTEX concentration,
was not contained in the docket and should be included and
considered by the EPA. The commenter indicated that since this
information was not referenced in the supplemental notice, it is
likely that the EPA did not give this data due consideration.
The commenter stated that the EPA needs to consider all available
data when establishing industry standards.
Response; The GRI report referred to by the commenters was
not submitted to the EPA, and therefore was not included in the
docket prior to the supplemental notice. However, since the
publication of the supplemental notice, the EPA has obtained a
copy of this report, and has reviewed its content (the report is
available on the Internet at the following address:
http://www.gri.org/pub/content/nov/19981105/115012/low-
btex_dehys.html) .
As stated in previous responses, the MACT floor for the
transmission and storage source category was developed based on
data collected in response to comments on the proposed
subpart HHH, as well as data collected prior to proposal. The
EPA determined the MACT floor for this source category to be
95-percent emission reduction. Furthermore, at least 93 percent
of the existing glycol dehydration units that are controlled, for
which the EPA had data (66 dehydration units out of 71 controlled
units) employed combustion in some form (i.e., flares, enclosed
combustion, or a combination of combustion and condensation) and
2-261
-------
.. "*""" ""
in response to these comments .
2-262
-------
1. REPORT NO.
EPA-453/R-99-004b
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
2.
4. TITLE AND SUBTITLE
National Emission Standards for Hazardous Air Pollutants for Source
Categories: Oil and Natural Gas Production and Natural Gas
Transmission and Storage - Background Information for Final
Standards. Summary of Public Comments and Responses
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Standards Division (Mail Drop 13)
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
12 SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
May 1999
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1 1 . CONTRACT/GRANT NO.
68-D6-0008
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16 ABSTRACT
This document contains a summary of public comments received on the NESHAP for Oil and Natural Gas
Production and Natural Gas Transmission and Storage (40 CFR 63, subparts HH and HHH), which were
proposed on February 6, 1998 (63 FR 6288). This document also provides the EPA's response to each
comment, and outlines the changes made to the regulation in response to public comments.
17
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b. IDENTIFIERS/OPEN ENDED TERMS c. COSATI
Field/Group
Environmental Protection, Air Pollution Control, Air Hazardous air pollutants
Emissions Control, Associated Equipment, Black Oil,
Condensate, Custody Transfer, Equipment Leaks,
Glycol Dehydration Units, Hazardous Air Pollutants,
Hazardous Substances, Natural Gas Transmission and
Storage, Oil and Natural Gas Production
Pipelines, Organic Liquids Distribution (non-gasoline),
Reporting and Recordkeeping Requirements, Storage
Vessels, Tank batteries, Tanks, Triethylene Glycol
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (Repon) 21 . NO. OF PAGES
Unclassified 285
20. SECURITY CLASS (Page) 22. PRICE
Unclassified
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |