United States - Environmental Protection Agency

Office of Air Quality Planning and Standards Research Triangle Park,
NC 27711

EPA-xxx/x-xx-xxx

December 2006

Air

National Emission Standards for Hazardous Air Pollutants:  Oil and
Natural Gas Production - Background Information for Final Standards

Summary of Public Comments and Responses



EPA-XXX/X-XX-XXX

National Emission Standards for Hazardous Air Pollutants:  Oil and
Natural Gas Production Facilities

Background Information for Promulgated Standards - Summary of Public
Comments and Responses.

Sector Policies and Programs Division

U.S. Environmental Protection Agency

Office of Air and Radiation

Office of Air Quality Planning and Standards

Research Triangle Park, NC  27711

TABLE OF CONTENTS

  TOC \o "1-4" \h \z \u    HYPERLINK \l "_Toc143490370"  1.0	SUMMARY	 
PAGEREF _Toc143490370 \h  1  

  HYPERLINK \l "_Toc143490371"  2.0	SUMMARY OF PUBLIC COMMENTS	  PAGEREF
_Toc143490371 \h  5  

  HYPERLINK \l "_Toc143490372"  2.1	General Comments on Proposal	 
PAGEREF _Toc143490372 \h  5  

  HYPERLINK \l "_Toc143490373"  2.1.1	Complexity	  PAGEREF _Toc143490373
\h  5  

  HYPERLINK \l "_Toc143490374"  2.1.2	Lag Time between Proposals	 
PAGEREF _Toc143490374 \h  6  

  HYPERLINK \l "_Toc143490375"  2.1.3	Request for Comment Extension	 
PAGEREF _Toc143490375 \h  6  

  HYPERLINK \l "_Toc143490376"  2.2	Compliance Date	  PAGEREF
_Toc143490376 \h  7  

  HYPERLINK \l "_Toc143490377"  2.3	Applicability	  PAGEREF
_Toc143490377 \h  10  

  HYPERLINK \l "_Toc143490378"  2.3.1	Applicability to TEG Units Only	 
PAGEREF _Toc143490378 \h  10  

  HYPERLINK \l "_Toc143490379"  2.3.2	Transmission of Oil and Natural
Gas	  PAGEREF _Toc143490379 \h  10  

  HYPERLINK \l "_Toc143490380"  2.3.3	Outer Continental Shelf	  PAGEREF
_Toc143490380 \h  11  

  HYPERLINK \l "_Toc143490381"  2.3.4	Geographic Applicability Criteria	
 PAGEREF _Toc143490381 \h  12  

  HYPERLINK \l "_Toc143490382"  2.3.5	Urban Definition	  PAGEREF
_Toc143490382 \h  17  

  HYPERLINK \l "_Toc143490383"  2.4	Exemptions	  PAGEREF _Toc143490383
\h  23  

  HYPERLINK \l "_Toc143490384"  2.4.1	Define Low-Risk Subcategory	 
PAGEREF _Toc143490384 \h  23  

  HYPERLINK \l "_Toc143490385"  2.4.2	Distance Exemption	  PAGEREF
_Toc143490385 \h  23  

  HYPERLINK \l "_Toc143490386"  2.4.3	Size Cutoffs	  PAGEREF
_Toc143490386 \h  25  

  HYPERLINK \l "_Toc143490387"  2.4.4	Decline Rates	  PAGEREF
_Toc143490387 \h  27  

  HYPERLINK \l "_Toc143490388"  2.4.5	Transition Policy On "Potential To
Emit"	  PAGEREF _Toc143490388 \h  28  

  HYPERLINK \l "_Toc143490389"  2.4.6	Documentation to Prove Exemptions	
 PAGEREF _Toc143490389 \h  29  

  HYPERLINK \l "_Toc143490390"  2.5	Control Requirements	  PAGEREF
_Toc143490390 \h  29  

  HYPERLINK \l "_Toc143490391"  2.5.1	Controls for Remote/Unmanned
Sources	  PAGEREF _Toc143490391 \h  29  

  HYPERLINK \l "_Toc143490392"  2.5.2	Start-up, Shutdown, Malfunction
Plans	  PAGEREF _Toc143490392 \h  31  

  HYPERLINK \l "_Toc143490393"  2.6	Reporting and Recordkeeping	 
PAGEREF _Toc143490393 \h  32  

  HYPERLINK \l "_Toc143490394"  2.7	Test Methods	  PAGEREF _Toc143490394
\h  33  

  HYPERLINK \l "_Toc143490395"  2.8	Impacts Analyses	  PAGEREF
_Toc143490395 \h  33  

  HYPERLINK \l "_Toc143490396"  2.8.1	TEG Dehydrator Population Estimate
  PAGEREF _Toc143490396 \h  33  

  HYPERLINK \l "_Toc143490397"  2.8.2	No New Area Source Assumption	 
PAGEREF _Toc143490397 \h  35  

  HYPERLINK \l "_Toc143490398"  2.8.3	Unmanned Facilities Use of Flares	
 PAGEREF _Toc143490398 \h  36  

  HYPERLINK \l "_Toc143490399"  2.8.4	Sensitivity To Operating Costs	 
PAGEREF _Toc143490399 \h  36  

  HYPERLINK \l "_Toc143490400"  2.9	Comments Received on February 6,
1998, Proposed Rule	  PAGEREF _Toc143490400 \h  37  

  HYPERLINK \l "_Toc143490401"  2.9.1	Definition of Urban Area	  PAGEREF
_Toc143490401 \h  42  

  HYPERLINK \l "_Toc143490402"  2.9.1.1	Interpretation of
Section 112(n)(4)	  PAGEREF _Toc143490402 \h  42  

  HYPERLINK \l "_Toc143490403"  2.9.1.2	Distinction between "Urban" and
"Urbanized"	  PAGEREF _Toc143490403 \h  43  

  HYPERLINK \l "_Toc143490404"  2.9.1.3	Classification of Entire
Counties as Urban	  PAGEREF _Toc143490404 \h  44  

  HYPERLINK \l "_Toc143490405"  2.9.1.4	Risk-distance Option	  PAGEREF
_Toc143490405 \h  46  

  HYPERLINK \l "_Toc143490406"  2.9.2	Area Source Regulation	  PAGEREF
_Toc143490406 \h  49  

  HYPERLINK \l "_Toc143490407"  2.9.3	Control Efficiency	  PAGEREF
_Toc143490407 \h  54  

  HYPERLINK \l "_Toc143490408"  2.9.4	Monitoring, Recordkeeping, and
Reporting	  PAGEREF _Toc143490408 \h  54  

  HYPERLINK \l "_Toc143490409"  3.0	REFERENCES	  PAGEREF _Toc143490409
\h  56  

 

SUMMARY

	The U.S. Environmental Protection Agency (EPA) placed in the Federal
Register on July 8, 2005 (70 FR 39441), a supplemental notice of
proposed rulemaking to the February 6, 1998 (63 FR 6288) proposed
national emissions standards for hazardous air pollutants (NESHAP) to
limit emissions of hazardous air pollutants (HAP) from oil and natural
gas production facilities that are area sources.  The final NESHAP for
major sources was promulgated on June 17, 1999 (64 FR 32610), but final
action with respect to area sources was deferred.  This July 8, 2005
action proposed changes to the 1998 proposed rule for area sources,
proposed alternative applicability criteria, and reopened the public
comment period to solicit comment on the changes proposed.  The public
comment period closed on September 6, 2005.  The purpose of this
document is to present a summary of the public comments received on the
supplemental proposal as well as the February 6, 1998 proposal and the
responses developed by EPA.

	The electronic docket for the July 8, 2005 supplemental proposal
contains 18 comment letters from 14 separate organizations.  The
commenters represent the following affiliations: industry (5 companies),
industrial trade associations (7 organizations), state agencies (1
agency), and an industry consultant (1 company).  Some of these
commenters supported by reference the substantive comments made by
others.  Table 1 presents a listing of all persons submitting written
comments, their affiliation, and the docket item number for their
comments.  The docket number for this action is EPA-HQ-OAR-2004-0238. 
Comments for the February 6, 1998 proposed rule are addressed in section
2.9 of this document (Comments Received on February 6, 1998, Proposed
Rule).Table 1.  List of Commenters on the Supplemental Notice of
Proposed Rulemaking on National Emission Standards for Hazardous Air
Pollutants:  Oil and Natural Gas Production Facilities

Document Docket ID	Commenter, Addressee, Title, or Description

0019	Stephanie R. Meadows

Upstream Coordinator

American Petroleum Institute

1220 L Street, NW

Washington, DC  20005

0020	Stephanie R. Meadows

Upstream Coordinator

American Petroleum Institute

1220 L Street, NW

Washington, DC  20005

0021	G.H.Holliday, Ph.D., P.E., DEE

Holliday Environmental Services, Inc. 

P.O. Box 2508

Bellaire, TX  77402

0022	Robert J Sandilos

Senior Government Relations Advisor

Chevron North America Exploration and Production

1500 Louisiana St.

Houston, TX  77002

0023	Lisa C. Moerner

Manager - Environmental Policy

Dominion Resources Services, Inc.

5000 Dominion Blvd

Glen Allen, VA  23060

0024	Pamela F. Faggart

Vice President and Chief Environmental Officer

Dominion Resources Services, Inc.

5000 Dominion Blvd

Glen Allen, VA  23060

0025	Margaret Young

Koch Exploration Company LLC

9777 Pyramid Court, Suite 210

Englewood, CO  80112

0026	Joel D. Howard

Manager - Health, Environment & Safety Services

Marathon Oil Company

P.O. Box 3128

Houston, TX  77253

0027	Gregory D. Russell

Vorys, Sater, Seymour, and Pease LLP

52 East Gay Street

Columbus, OH  43216

on behalf of the Ohio Oil and Gas Association

0028	Larry Lashley

Louisiana Department of Environmental Quality

318-676-5088

0029	Robert Radabaugh

Chairman

Independent Oil and Gas Association of West Virginia

405 Capitol Street, Suite 507

Charleston, WV  25301

0030	Barry Russell

President

Independent Petroleum Associaton of America

0031	Angie Burckhalter

V.P. of Regulatory Affairs

Oklahoma Independent Petroleum Association

3555 N.W. 58th Street, Suite 400

Oklahoma City, OK  73112

0032	Lisa Beal

Director

Environment and Construction Policy

Interstate Natural Gas Association of America

10 G Street, N.E., Suite 700

Washington, DC  20002

0033	Lisa Beal

Director

Environment and Construction Policy

Interstate Natural Gas Association of America

10 G Street, N.E., Suite 700

Washington, DC  20002

0034	Nicholas DeMarco

Executive Director

West Virginia Oil and Natural Gas Association

P.O. Box 3231

Charleston, WV  25332

0035	Stephanie R. Meadows

Upstream Coordinator

American Petroleum Institute

1220 L Street, NW

Washington, DC  20005

0036	Stephanie R. Meadows

Upstream Coordinator

American Petroleum Institute

1220 L Street, NW

Washington, DC  20005

0043	Don Scott Wallace

Sr. Environmental Specialist

Devon Energy Corporation

20 North Broadway

Oklahoma City, Oklahoma 73102



SUMMARY OF PUBLIC COMMENTS

General Comments on Proposal

Complexity

	Comment:  One commenter (0021) stated that the proposal was too
complex.  The commenter (0021) stressed that many TEG dehydrators are
operated by small independent Exploration and Production (E&P)
operators.  The commenter (0021) was concerned that the format of the
proposal (i.e., amendatory text) precluded small independent operators
from providing comments.   The commenter (0021) maintained that
cutting/pasting the amendatory language into subpart HH is a daunting
task not only for an experienced person, but it would be an impossible
task for an independent E&P operator who is not subject to the major
source rule.  The commenter (0021) recommended that EPA withdraw the
proposed rule and prepare a more understandable rule for only area
sources.

	Response: While this is our standard method of amending rules that
have been published in the code of federal regulations (CFR), we agree
with the commenter that reading this amendatory language may be
difficult.  Therefore, we have added a copy of the entire rule (in
redline/strikeout format) to the oil and natural gas production website
(  HYPERLINK "http://www.epa.gov/ttn/atw/oilgas/oilgaspg.html" 
http://www.epa.gov/ttn/atw/oilgas/oilgaspg.html ).  Since the control
requirements for area source TEG dehydrators are similar to the
requirements for major source dehydrators contained in 40 CFR part 63
subpart HH, we believe it is more efficient to amend subpart HH to
incorporate area source provisions rather than create a separate area
source rule as suggested by the commenter.

	Comment:  One commenter (0021) referred to the applicability options
for the location criteria.  Specifically, the commenter (0021) stated
that by offering two options for area source control implementation when
an area source changes from Rural to Urban-1 or Urban-2 status, EPA has
increased the difficulty of understanding the proposed rule.  The
commenter (0021) also referred to EPA's using census data from 2010 and
later as a criterion for the conversion of rural to urban dehydrators as
adding complexity.

	Response:  We recognize that proposing two options with different
compliance dates for rural sources made the proposed rule more
complicated to evaluate, however, this was the only approach that
allowed us to provide the same compliance period for sources regulated
under the two options (i.e., three years).  We believe that it is
necessary to use the most current data from the Bureau of Census since
it recognizes where people are located.  We plan to use notice and
comment rulemaking through the Federal Register to amend the rule’s
applicability criteria to reflect the latest decennial Census data.

Lag Time between Proposals

	Comment:  Commenter 0031 asked why there has been such a long lag time
between the original 1998 proposal and the 2005 proposal.  Has EPA
evaluated current data and information that justifies the need for
additional requirements on upstream oil and gas production TEG units?

	Response:  As stated in the July 8, 2005, preamble (70 FR 39443), the
basis for regulating area source oil and natural gas production
facilities shifted from the 1998 proposed rule's area source finding (a
proposed finding of adverse human health effects from benzene emissions
from TEG dehydration units) to the 2005 proposed rule which fulfills a
portion of our obligation under section 112(c)(3) to regulate stationary
sources of benzene that are area sources.  We deferred final action on
the 1998 proposed rule pending issuance on the Urban Air Toxics Strategy
(UATS).  Although the UATS was finalized in July 1999 (64 FR 38706), we
set our future plans based on court-ordered deadlines.  The court
ordered deadline for regulating area source oil and gas production
facilities is December 2006.

Request for Comment Extension

	Comment:  Three commenters (0019, 0020, 0031) requested an extension to
the public comment period.  Two commenters (0019, 0020) requested a
30-day extension in order to thoroughly evaluate the notice and provide
helpful comments to EPA.  The third commenter requested a 60-day
extension so that smaller oil and gas operators could evaluate the
impacts of the proposed rule on their oil and gas sites.

	Response:    SEQ CHAPTER \h \r 1 We denied requests to extend the
public comment period.  Given that we are beyond the statutory deadline
for promulgating the area source NESHAP and because this is a supplement
to the February 6, 1998, proposed rule, we believe that 60 days was a
reasonable public comment period.

	Further, because the proposed rule potentially applied on a national
basis, we believe that 60 days was an adequate period of time to
determine whether or not a facility has a TEG dehydration unit and
whether the unit operates above the 3-MMscf/day and 1-tpy benzene
cutoffs and to draft comments.

Compliance Date

	Comment:  Several commenters (0022, 0023/0024, 0025, 0026, 0030, 0031,
0032/0033, 0034, 0035/0036. 0043) requested that EPA modify the
compliance date.  One commenter (0023/0024) suggested that the date used
to differentiate new from existing dehydrators should be July 8, 2005. 
The commenter (0023/0024) stated that EPA's reasoning for using February
6, 1998 to differentiate between existing and new sources was that the
July 8, 2005 proposed rule is a supplement to the 1998 proposed rule,
which had a new source threshold date of February 6, 1998.  However,
when the final rule was published on June 17, 1999, it did not regulate
area sources.  Therefore, according to the commenter (0023/0024), it is
unreasonable for dehydrators installed between February 6, 1998 and July
8, 2005 to be treated as new dehydrators in the area source rule.  

	Five commenters (0022, 0026, 0030, 0035/0036, 0043) requested that the
compliance date be adjusted to accommodate the delay between the
original proposal and the supplemental proposal and because of confusion
in the applicability determinations in the proposed rule.  The
commenters (0022, 0026, 0030, 0035/0036) argued that the 1998 proposal
did not clearly communicate the proposed geographic scope of the
proposed area source standard.  According to the commenters (0022, 0026,
0030, 0035/0036),  the definition of Urban-1 was clear, but because the
Urban-2 definition was unclear, E&P operators would not be able to
adequately determine whether the standard would apply outside of Urban-1
areas.  Therefore, the commenters (0022, 0026, 0030, 0035/0036) argued
that due to the delay between the two proposals, and the significant
confusion in the applicability of the 1998 proposal, it was unreasonable
for EPA to adopt a schedule for implementing the area source rule since
E&P operators were not provided reasonable notice regarding area source
control requirements outside of Urban-1 areas.  The commenters (0022,
0026, 0030, 0035/0036) recommended that EPA grant area sources outside
of Urban-1 counties three years to achieve compliance if their
construction or reconstruction began after the date of the 1998 proposal
(i.e., treat them as existing sources).  The commenters (0022, 0026,
0030, 0035/0036) suggested the following revision to proposed paragraphs
(f)(3) through (6) to accommodate that change: 

	(f) The owner or operator of an affected major source shall achieve
compliance with the provisions of this subpart by the dates specified in
paragraphs (f)(1)and (2) of this section. The owner or operator of an
affected area source shall achieve compliance with the provisions of
this subpart by the dates specified in paragraphs (f)(3) through (6) of
this section.

	(1) The owner or operator of an affected major source, the construction
or reconstruction of which commenced before February 6, 1998, shall
achieve compliance with the applicable provisions of this subpart no
later than June 17, 2002 except as provided for in § 63.6(i).

* * *

	(2) The owner or operator of an affected major source, the construction
or reconstruction of which commences on or after February 6, 1998, shall
achieve compliance with the applicable provisions of this subpart
immediately upon initial startup or June 17, 1999, whichever date is
later.

* * *

	(3) The owner or operator of an affected area source located in an
urban area, as defined in § 63.761, the construction or reconstruction
of which commences before February 6, 1998, shall achieve compliance
with the provisions of this subpart no later than 3 years after the date
of publication of the final rule in the Federal Register except as
provided for in § 63.6(i).

	(4) The owner or operator of an affected area source located an area
classified "Urban-1" in an urban area, as defined in § 63.761, 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 date of publication of the final
rule in the Federal Register, whichever date is later.

	(5) The owner or operator of an affected area source located in rura an
area classified "Urban-2" in an urban area, as defined in § 63.761, the
construction or reconstruction of which commences before July 8, 2005
shall achieve compliance with the provisions of this subpart no later
than 3 years after the date of publication of the final rule in the
Federal Register except as provided for in § 63.6(i).

	(6) The owner or operator of an affected area source located in a
ruralan area classified "Urban-2" in an urban area, as defined in §
63.761, the construction or reconstruction of which commences on or
after July 8, 2005 shall achieve compliance with the provisions of this
subpart immediately upon initial startup or date of publication of the
final rule in the Federal Register, whichever date is later.

	The commenters (0022, 0026, 0030, 0035/0036) further stated that by
adopting a standard that imposes controls  where none were anticipated,
EPA would be imposing controls that are more stringent than its proposed
controls.  The commenters (0022, 0026, 0030, 0035/0036) stated that they
believe that the CAA authorizes EPA to defer the compliance date for
area sources outside of Urban-1 areas, because the final rule would be
more stringent than the proposed rule.  According to the commenters
(0022, 0026, 0030, 0035/0036) 40 CFR §63.6(b)(3) provides EPA the
authority to adjust the implementation schedule.

	Response:  We agree with the commenters that the confusion related to
the definition merits a change in how we classify existing and new
sources, which then changes the compliance date for certain sources.  We
believe that area sources not located in counties classified as Urban-1
should be considered existing sources if constructed/reconstructed
before July 8, 2005.   Therefore, the final rule defines existing and
new sources as follows:

Sources located in counties classified as Urban-1 based on the 1990
Census and were constructed or reconstructed before February 6, 1998 are
considered existing sources.

Sources that are not located in a county classified as Urban-1 based on
the 1990 Census and were constructed or reconstructed before July 8,
2005 are considered existing sources.

Sources located in a county classified as Urban-1 based on the 1990
Census and were constructed on or after February 6, 1998 are considered
new sources.

All sources constructed or reconstructed on or after July 8, 2005 are
considered new sources.

In addition, since the final rule requires add-on controls for certain
sources and management practices at others, the compliance period
differs based on the emission reduction requirements for existing
sources [discussed in more detail in section 2.5 of this document
(Control Requirements)].  Existing sources installing add-on controls
must achieve compliance no later than three years after the final
rule’s effective date in the Federal Register.  Existing sources
implementing management practices must achieve compliance no later than
one year after the effective date.

The commenters’ statements regarding 40 CFR 63.6(b)(3) are not
relevant since the July 8, 2005 supplemental notice informed all area
sources that they could be required to install add-on controls, and
therefore the final rule is not more stringent that the proposed rule.

Applicability

Applicability to TEG Units Only

	Comment:  One commenter (0025) supported EPA's conclusion that the
standards should only apply to TEG dehydration units.  They stated that
this is the only approach authorized at this time based on the fact that
oil and gas production facilities were listed as an area source under
the Urban Air Toxics Strategy solely because TEG dehydration units were
deemed to be a significant source of benzene, and benzene is one of the
30 pollutants regulated under this strategy.  Therefore, the commenter
(0025) concluded that these TEG dehydration units are the only units
that can be subject to control requirements under this strategy.

	Response:  As stated in the preamble to the supplemental proposal (70
FR 39443), oil and natural gas production facilities were listed in the
Urban Air Toxics Strategy solely because the TEG dehydration units
located at these facilities contributed approximately 47 percent of the
national urban emissions of benzene from stationary sources at area
sources.  We continue to believe that it is appropriate to define the
affected source as each TEG dehydration unit located at oil and natural
gas production facilities and we have not changed the definition of
affected source in the final rule.

Transmission of Oil and Natural Gas

	Comment:  Commenters (0032/0033, 0034) seek confirmation of the
applicability of these rules to only production units, not transmission
of oil and natural gas.  The table on page 39442 refers to units with
NAICS Codes 211111 and 211112, but also directs one to the Subpart HH
applicability criteria.  The commenters requested a clear exclusion of
transmission facilities so that no mistakes will be made by regulated
entities.

	Response:  The commenters concerns about the applicability of subpart
HH to transmission and storage units are not clear.  The NAICS Code for
natural gas transmission and storage facilities is 486210.  Both the
NAICS Codes listed in the table on page 39442 of the preamble to the
supplemental proposed rule (NAICS Codes 211111 and 211112) and subpart
HH refer to oil and natural gas production facilities.

	Further, we do not believe that a specific exclusion for natural gas
transmission and storage facilities is necessary.  As we discussed in
our response to a similar comment on the 1998 proposed rule, we believe
that the definitions in subparts HH (§63.761) and HHH (§63.1271)
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 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.  Applicability to the area source rule for
oil and natural gas production facilities is specified in §63.760(a),
which states that "facilities that process, upgrade, or store natural
gas prior to the point at which natural gas enters the natural gas
transmission and storage source category or is delivered to a final end
user" are subject to subpart HH [§63.760(a)(3)].  Section 63.760(a)(3)
further specifies that "for the purposes of [subpart HH], natural gas
enters the natural gas transmission and storage source category after
the natural gas processing plant, when present.  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."  Therefore, based on this applicability criterion, no
specific exemptions for natural gas transmission and storage facilities
have been added to subpart HH.  

Outer Continental Shelf

	Comment:  Commenters (0022, 0026, 0030, 0035/0036) said EPA should
confirm that the area source proposal is not intended to apply to Outer
Continental Shelf (OCS) facilities.  Commenters disagreed with EPA's
statement in the proposal that these sources should be excluded solely
because they are assumed to already have controls.  Instead, EPA should
also recognize that the authority for controlling emissions from
facilities in the OCS in the central and western Gulf of Mexico is
specifically reserved to the Secretary of the Interior and that in areas
where EPA does have authority over OCS sources, it only extends to
emissions of criteria pollutants, not HAP.  Excluding OCS sources from
area source controls conforms to these precedents.

	Response:  This area source rule applies to OCS facilities.  As we
stated in our response to comments received on OCS for the major source
rule, section 328 of the CAA provides us the authority to regulate OCS
sources to attain and maintain Federal and state ambient air quality
standards and to comply with the provisions of the CAA title I, part C. 
  EPA’s implementing regulation at 40 CFR § 55.13(e) specifies that
“provisions promulgated pursuant to section 112 of the Act shall apply
[to OCS sources] if rationally related to the attainment and maintenance
of federal or state ambient air quality standards ....”  This area
source rule regulates benzene, a volatile organic compound (VOC).  VOCs
react with nitrogen oxides to form ozone, one of the criteria pollutants
for which EPA is required to set national air ambient air quality
standards.  Accordingly, this rule applies to OCS sources because its
regulation of benzene is rationally related to the "attainment and
maintenance" of ambient air quality standards for ozone. 

	Our statement that none of the sources on the OCS are uncontrolled area
sources that would be impacted by the final rule was in reference to the
assumptions that were made to estimate impacts and was not intended to
imply that this area source rule does not apply to OCS sources.  The
commenters did not provide additional data contradicting our assumption
that offshore platforms located on the OCS are generally controlled by a
flare for safety purposes.  Therefore, our assumption and impact
analysis remain unchanged.

Geographic Applicability Criteria

Comment:  EPA proposed two options for the geographic applicability
criteria: (1) all TEG dehydration units would be subject to area source
standards (referred to throughout this document as Option 1) ; and (2)
area source standards would apply to TEG dehydration units located in
Urban-1 and Urban-2 counties (referred to throughout the remainder of
this document as Option 2).  EPA specifically requested comments on
Options 1 and 2.  Fourteen commenters (0021, 0022, 0023/0024, 0025,
0026, 0027, 0028, 0029, 0030, 0031, 0032/0033, 0034, 0035/0036, 0043)
responded to EPA's request for comments.  

	One commenter (0028) supported Option 1.  In support of Option 1, the
commenter (0028) stated that rules should not be different for
facilities located in urban counties than those for facilities located
in rural counties.  According to the commenter, who works with small
businesses, a common theme was that small businesses want a level
playing field.  The commenter (0028) further stated that it puts
additional burden on environmental inspectors if different rules are
applicable in different counties.

Twelve commenters (0022, 0023/0024, 0025, 0026, 0027, 0029, 0030, 0031,
0032/0033, 0034, 0035/0036, 0043) were opposed to Option 1.  The
commenters were opposed to EPA's imposing control requirements on area
sources in rural areas for two primary reasons: (1)  EPA did not have
the authority to regulate rural sources and (2)  low exposure risks
from rural or remote sources.  Several of the commenters (0023/0024,
0026, 0029, 0030, 0032/0033, 0034, 0035/0036, 0043) supported Option 2,
with changes to the definition of urban areas (see section 2.3.5 for
specific comments regarding the definition of urban areas). 

Several commenters (0022, 0025, 0026, 0029, 0030, 0031, 0032/0033, 0034,
0035/0036, 0043) stated that EPA did not have the authority under the
Clean Air Act (CAA) to regulate area sources in rural areas.  One
commenter (0043) stated that Option 2 better meets the intent of the
area source rules.  Two commenters (0032/0033, 0034) stated that
nationwide applicability is contrary to the plain language of the CAA
which intends for the area source to address HAP in urban areas.  Six
commenters (0022, 0026, 0029, 0030, 0031, 0035/0036) agreed that the
purpose of the area source program was to reduce "...ambient
concentrations [of HAP] characteristic of large urban areas to levels
substantially below those currently experienced."

	Several commenters (0022, 0023/0024, 0026, 0029, 0030, 0035/0036),
specifically referred to section 112(k) of the CAA which, according to
the commenters, makes it apparent that the concern to be addressed was
unique to urban areas.  One commenter (0023/0024) stated that regulating
rural dehydrators provides limited health benefit since populations in
these areas are sparse.  According to the commenter (0023/0024), the
impact of benzene emissions beyond facility fencelines would be minimal
since the number of other sources of benzene emissions are few.  In
addition, the commenter stated that many States have air toxics
regulations which limit the emissions of HAP based on impact on the
health of populations beyond the fenceline.  Three commenters (0022,
0023/0024, 0029) further stated that it is clear that the remote, small,
or sparsely populated rural areas which characterize the majority of the
dehydrators potentially subject to the rural area applicability option
are not within the scope of section 112(k)(1).  One commenter (0025)
referred to EPA's decision to regulate area sources based on the fact
that the oil and natural gas production source category was listed in
the Urban Air Toxics Strategy, which focuses on urban air toxics
emissions as directed by section 112(k)(3)(A) and that nothing in the
CAA authorizes extending the Urban Air Toxics program to rural areas. 
The commenter further stated that even if section 112(k)(3) did allow
regulation of area sources in rural counties, EPA has not made the
necessary findings to justify extending the proposed rule to rural
locations.  

Several commenters (0022, 0025, 0026, 0029, 0030, 0031, 0035/0036, 0043)
referred to exposure risks from facilities located in rural or remote
areas.  One commenter (0031) stated that exposure risks in remote areas
are low or nonexistent.  Two commenters (0032/0033 and 0034) stated that
there is no clear indication that emissions from remote sources provide
a meaningful contribution to ambient air toxic levels in urban areas. 
One commenter (0029) stressed that the foundation for the area source
program was based on regulating area sources in a manner that would
result in a public health benefit.  The commenter stated that regulating
dehydration units in rural areas would not yield the same public health
benefits that were "contemplated" by the statute.   

Five commenters (0022, 0026, 0030, 0034, 0035/0036) stated that
regulating rural sources would not have the effect intended by the CAA. 
The commenters (0022, 0026, 0030, 0034, 0035/0036) cited section 112(c)
of the CAA, where Congress directed EPA to list for controls "...area
source categories representing 90 percent of the area source emissions
of the 30 hazardous air pollutants that present the greatest threat to
public health in the largest number of urban areas."  According to the
commenters, the focus of the area source program is those smaller
sources of HAP that create unacceptable exposures in concentrated urban
areas.  According to one commenter (0034), the purpose of regulating
area sources under section 112 of the CAA was to specifically reduce
health risks to the environment and people in urban areas (i.e.,
considered to be densely populated).  Four commenters (0022, 0026, 0030,
0035/0036) stated that most of EPA's estimated 38,000 dehydrators in the
E&P sector are in remote, rural locations and do not present the risk of
exposure that the area source controls are meant to address.  

Two commenters (0031, 0043) stated that regulating area sources under
Option 1 was unnecessarily costly and burdensome on sites located in
these remote areas and had little or no effect on human health.

Response:  We believe that the CAA provides the Agency with the
authority to regulate area sources nationwide.  CAA section 112(k)(1)
states that “It is the purpose of this subsection to achieve a
substantial reduction in emissions of hazardous air pollutants from area
sources and an equivalent reduction in the public health risks
associated with such sources including a reduction of not less than 75
per centum in the incidence of cancer attributable to emissions from
such sources.”  Consistent with this expressed purpose of CAA section
112(k) to reduce both emissions and risks, CAA section 112(k)(3)(i)
requires that we list not less than 30 HAP that, as a result of
emissions from area sources, present the greatest threat to public
health in the largest number of urban areas.  CAA sections 112(c)(3) and
(k)(3)(ii) require that we list area source categories that represent
not less than 90 percent of the area source emissions of each of the
listed HAP.  CAA section 112(c) requires that we issue standards for
listed categories under CAA section 112(d).  These relevant statutory
provisions authorize us to regulate listed area source categories and
not just sources located in urban areas. 

	In both the UATS and our July 8, 2005 supplemental proposal, we
identified the reasons supporting a national rule (e.g., benzene’s
toxicity and carcinogenicity, a level playing field, the 75 percent
cancer incidence reduction goal) (64 FR 38724 and 70 FR 39446). 
Furthermore, by requiring management practices rather than control
requirements on sources outside the UA  plus offset and UC boundary, we
believe that we have appropriately addressed commenters’ concern with
respect to remote sources being subject to unnecessary or costly
requirements.

Comment:  Several commenters (0022, 0026, 0030, 0035/0036) stated that
extending area source controls to those remote, rural sources far
exceeds the Congressional mandate to control potential threats to public
health in urban areas from diverse sources of HAP emissions that do not
qualify as major sources, as expressed in sections 112(c), 112(k), and
112(n)(4)(B) of the CAA (42 U.S.C. §§7412(c), 7412(k), and
7412(n)(4)(B)).  The commenters further stated that Congress clearly
recognized that rural E&P operations presented a low risk of public
exposure to HAP.  In support of their argument, the commenters pointed
to section 112(n)(4)(A) and (B), where the commenters stated that
Congress provided express direction to EPA on the treatment of E&P
operations.  According to the commenters, based on these sections in the
CAA, it was clear that Congress recognized that the remote, rural nature
of most E&P operations presented a decreased risk of public exposure to
HAP emissions.  The commenters stated that it prohibited the aggregation
of emissions from similar equipment in a contiguous area, and it limited
the listing of E&P area sources outside of truly urban areas.  

Response: As stated in a previous response, we believe that sections
112(c) and 112(k) do not prohibit us from issuing area source rules of
national applicability.  We also disagree with the commenters' statement
that sections 112(n)(4)(A) and (B) of the CAA limits the listing of E&P
area sources outside truly urban areas.  First, section 112(n)(4)(A)
does not address area sources.  Secondly, section 112(n)(4)(B) states
that we "...shall not list oil and gas production wells (with its
associated equipment) as an area source category...except that (we) may
establish an area source category for oil and gas production wells
located in any metropolitan statistical area or consolidated
metropolitan statistical area with a population in excess of 1
million...."  In the February 1998 proposal preamble, we addressed the
definition of "associated equipment" and determined that glycol
dehydration units and storage vessels with flash emissions are not
associated equipment (63 FR 6300).  Therefore, section 112(n)(4)(B) does
not apply to TEG dehydration units at oil and natural gas production
facilities.

	Comment:  Several commenters (0022, 0026, 0030, 0035/0036) stated that
EPA did not suggest in the 1998 proposal (63 FR 6300, 6306, and 6309)
that it was considering imposing area source controls on rural TEG
dehydrators.

	Response:  The commenter is correct that we did not indicate in the
1998 proposal that we were considering imposing area source controls on
rural TEG dehydrators.  It is for this reason (among others) that we
published the July 2005 supplemental proposed rule and accepted comments
on Option 1.

	Comment:  One commenter (0021) stated that they felt that Option 1
(i.e., regulating all TEG dehydration units) is more favorable to
operators.  Regarding Option 2, the commenter stated that it presented a
technically correct method of dealing with rural areas as they mature
into urban areas with time, but that it imposes more severe time
constraints.  According to the commenter (0021), Option 2 mandates:

"pre 6 February 1998" rural area sources comply with 40 CFR subpart HH
within three years of the publication of the rule;

"pre 6 February 1998" area sources which later become rural as the
result of census update comply with subpart HH within three years of
publication of the census update; and

"post 5 February 1998" rural area sources comply with subpart HH upon
publication of the rule.

The commenter stated that this would result in many "post 5 February
1998" facilities being forced to "shut in" while locating and installing
controls.

	Response:  The compliance schedule for new sources is dictated by the
statute (i.e., section 112(i) of the CAA).  While new sources must
comply with the final rule on the date of publication of the final rule
or upon startup (whichever is later), the final rule specifies that
sources not located in Urban-1 counties are only considered new sources
if they were constructed on or after July 8, 2005.  Therefore, fewer
sources constructed after February 1998 are considered new sources in
the final rule.

Comment:  One commenter (0025) pointed out that the proposal did not
include an evaluation of the hazards TEG units pose in rural areas, and,
according to the commenter, by definition rural areas involve much lower
exposure risks than urban areas.  The commenter noted that in the 1998
proposal, EPA was careful to quantify the cancer risks that it believed
were posed by area source TEG units (63 FR 6288, 6299) but EPA did not
update these figures in the July 2005 proposal, instead relying on
"general" statements to justify extending the rule to rural areas (70 FR
39446).  According to the commenter (0025), EPA's general statements did
not provide an adequate legal basis for extending control requirements
to rural areas.

	Response:  As stated in the July 8, 2005, preamble to the supplemental
proposal (70 FR 39443), the February 6, 1998 proposed area source rule
was based on a proposed finding of adverse human health effects from TEG
dehydration units at area source oil and natural gas production
facilities.  The basis for the July 2005 supplemental proposal and the
final rule is to fulfill a portion of our obligation under sections
112(c)(3) and 112(k)(3)(B) to regulate area source categories accounting
for 90 per centum or more of emissions of 30 identified HAP that present
the greatest threat to public health in the largest number of urban
areas (“area source HAP”).  Accordingly, listing of area source
categories under these provisions was based on the categories’
contributions to area source HAP emissions and not on a risk finding. 
TEG dehydration units at oil and gas production facilities were listed
because they contributed significantly to emission of benzene, one of
the 30 area source HAP.

Urban Definition

	Comment:  Several commenters (0022, 0024, 0026, 0027, 0029, 0030, 0031,
0032/0033, 0034, 0035/0056) opposed EPA's definition of "urban areas." 
According to the commenters (0022, 0024, 0026, 0029, 0030, 0031,
0032/0033, 0034, 0035/0056), by defining urban area as county-wide
areas, EPA has expanded urban areas to include large expanses of rural
territories.  One commenter (0034) stated that a comparison of land area
to population on a county basis shows that the target population for
protection is very thinly distributed.  Four commenters (0022, 0026,
0030, 0035/0056) referred to maps (included in comment 0036 as Exhibit
A).  The commenters noted that the maps show vast areas of the United
States that would be classified as urban areas based on the proposed
definition, but have very low population.  The commenters specifically
referred to the State of Wyoming, in which half of the State is
classified as "urban" using EPA's proposed definition.  One commenter
(0026) also pointed out that in Utah, six of the twelve counties
designated as "urban" using EPA's definition have a population density
of less than ten persons per square mile.

Other commenters (0032/0033, 0034) stated that some counties with a
total population of less than 5,000 and an average population density of
less than 2 people per square mile would be classified as urban under
the Urban-2 designation.  In order to illustrate the broad geographical
applicability that includes remote locations, the commenters stated that
based on the Urban-2 definition, urban designations would be applied to:

14 of 23 counties in Wyoming;

20 of 33 counties in New Mexico;

10 or 17 counties in Nevada; and 

17 of 56 counties in Montana.

	One commenter (0031) stated that EPA's proposed definition of urban
areas would be unnecessarily costly and burdensome on sites located in
rural or remote areas but classified as urban.

Three commenters (0031, 0032/0033, 0034) maintained that EPA does not
have the authority under section 112(k) to regulate area sources in
non-urban, rural areas.  One commenter (0031) stated that regulating
these sources is outside the scope of the law.  Two commenters
(0032/0033, 0034) stated that by classifying many geographical areas as
urban areas is beyond a reasonable interpretation of the intent of
section 112(k) of the CAA.  The commenters argued that the intent of
section 112(k) was to regulate "urban" or "large urban areas" and that a
number of areas that are included are not "urban" or "large urban
areas."  The commenters stated that it is apparent that the intent of
the CAA and the Urban Strategy is to affect change in urban areas and
EPA should identify the rationale for including remote areas in the
standard.

One commenter (0022) acknowledged that there have been and will continue
to be instances of energy production and population encroachment. 
However, according to the commenter, most of the known conventional or
unconventional gas supply basins are likely to remain rural for the
foreseeable future.

Response:  The statute does not define urban, thus, leaving us the
discretion to define the term.  We proposed and took comments on our
definition of the term urban as part of our 1999 UATS.  The definition
was the basis for the listing of area source categories pursuant to
section 112(c)(3) and (k)(3)(B)(ii) of the CAA.  We are currently under
court-ordered deadlines to complete issuing standards for all listed
area source categories.  Changing the definition of urban would mean
recreating an area source category list, which may differ significantly
from the current list and, thus, greatly hinders our effort to complete
our obligation by the court-ordered deadlines.  Therefore, we believe
that revisiting the definition of urban is inappropriate at this time. 
However, we have tailored this rule to address the unique circumstances
associated with this source category, as described above.  Moreover, in
response to comments regarding the nature of remote sources, we modified
this final rule and are only requiring the add-on control requirement
for sources in areas of higher population densities, which we have
identified as areas within the UA plus offset and UC boundaries.  This
rule imposes the less costly management practice requirements on sources
outside the UA plus offset and UC boundaries.  

	Comment:  Several commenters (0022, 0024, 0026, 0027, 0029, 0030,
0032/0033, 0034, 0035/0056) recommended that EPA redefine "urban areas."
 Two commenters (0029, 0034) suggested that the definition of urban area
should at most be limited to Urban-1 counties based on the 2000 census. 
According to one of the commenters (0034), emission reductions would not
have the effect intended by and directed through the CAA, but would
instead be reductions in open countryside where few, if any, will have a
benefit.

Two commenters (0032/0033, 0034) stated that EPA should not be compelled
to use the Census Bureau-based definition of urban areas and that other
viable options are available.  The commenters (0032/0033, 0034) stated
that alternative definitions are available from other agencies and from
recent Federal Register releases. The commenters (0032/0033, 0034)
stated that they believe that EPA should revise the definition such that
it does not include remote or sparsely populated regions.  According to
the commenters (0032/0033, 0034), several notices have been published in
the Federal Register related to defining "urban" versus "rural" and
Metropolitan Statistical Area (MSA).  As an example, the commenters
(0032/0033, 0034) referred to one notice which indicates that there are
at least six definitions of "urban" in use within federal agencies (65
FR 82229).  The commenters (0032/0033, 0034) recommended that EPA define
an alternative option for geographic applicability that does not use the
Urban-2 definition and considers 2000 census information.  For example,
the commenters referred to "Urban Area Criteria for Census 2000" (67 FR
11663) which includes the following definition:

"For Census 2000, a UA [urbanized area] consists of contiguous, densely
settled census block groups (BGs) and census blocks that meet minimum
population density requirements, along with adjacent densely settled
census blocks that together encompass a population of at least 50,000
people."

The commenters also referred to the Census Bureau's online glossary
definition of "urbanized area:"

"An area consisting of a central place(s) and adjacent territory with a
general population density of at least 1,000 people per square mile of
land area that together have a minimum residential population of at
least 50,000 people.  The Census Bureau uses published criteria to
determine the qualification and boundaries of UAs."

According to the commenters, the Census Bureau published a list of 453
urbanized areas in the U.S. (based on the 2000 census) with populations
over 50,000, which comprise about 70 percent of the total U.S.
population (67 FR 21962).  The commenters stated that the areas and
population covered using this definition appear more than adequate to
address the intent of section 112(k) of the CAA (i.e., to reduce ambient
levels of air toxics for urban areas and populations).

The two commenters (0032/0033, 0034) also referenced the definition of
"urban cluster," a densely settled territory that has at least 2,500
people but fewer than 50,000, which delineates urban and rural areas by
census blocks.  The commenters noted that census blocks are much smaller
than counties and provide a more precise urban-rural partitioning than
counties.  According to the commenters (0032/0033, 0034), using this
definition would add to the urban category over 3,000 additional areas
and about an additional 10 percent of the U.S. population.  The
commenters (0032/0033, 0034) also stated that including the "urban
cluster" definition could add complexity and extend beyond highly
populated, larger urban areas.  The commenters (0032/0033, 0034) did
state that including Core Based Areas, Urban Clusters, and Micropolitan
Areas (which are detailed in the Census Bureau releases) did not seem
necessary to address the intent of the area source program.  

Five commenters (0022, 0024, 0026, 0030, 0035/0056) provided an
alternative definition for urban areas which includes Census-defined MSA
and "urbanized areas," as follows:

Urban area for the purposes of the area source determination is defined
by use of the U.S. Department of Commerce's Bureau of Census statistical
data to classify all land area in the U.S. into one of three
classifications as follows:

Urban-1 areas, which consist of metropolitan statistical areas (MSA)
with a population greater than 250,000.

Urban-2 areas, which are defined as all other areas designated as
"urbanized" by the Bureau of Census (areas which comprise one or more
central places and the adjacent densely settled surrounding fringe that
together have a minimum of 50,000 persons.  The urban fringe consists of
contiguous territory having a density of at least 1,000 persons per
square mile); or 

Rural areas, which are all areas that are not designated as Urban-1 or
Urban-2.

The five commenters (0022, 0024, 0026, 0030, 0035/0056) claimed that by
revising the definition, EPA would create a more appropriate focus for
applying the regulation.  According to the commenters, their definition
more closely approximates the target cited in the Integrated Urban
Strategy, which was urban areas with "populations of more than 50,000." 
The commenters further argued that their recommended definition would
more clearly implement the purpose of the area source program, to
address HAP exposure in urban areas.

	Two commenters (0032/0033, 0034) supported the use of the "urbanized
area" definition as providing an appropriate basis to identify areas
that should be regulated under the proposed standards and is consistent
with the intent of the CAA.  At a minimum, the commenters (0032/0033,
0034) stated that EPA should evaluate the range of alternatives that are
available from other Federal agencies (e.g., urbanized area, urban
cluster) to redefine the urban area definition. 

	One commenter (0027) suggested that EPA redefine urban area or consider
other mechanisms for determining appropriate population densities based
on the risks presented.  For example, the commenter suggested a
mechanism analogous to the U.S. Department of Transportation's use of
the Potential Impact Circle concept when regulating high pressure
natural gas transmission lines for purposes of pipeline safety.

 	Several commenters (0022, 0026, 0030, 0035/0056) also pointed out that
EPA noted in the Integrated Urban Strategy states that "the
determination of the area in which standards apply should be made
separately for each source category."  According to the commenters, EPA
does not need to impose uniform definitions of urban areas for all area
sources.

	Response:  As previously stated, our long-standing definition of
“urban” is the product of public notice and comment and the
foundation of the Agency’s area source program.  Because the
definition applies to all listed area source categories, we do not
believe that it is appropriate to change the definition within any
specific area source rule.  However, we would find ways to address any
specific concern with the application of the definition when warranted. 
In this case, because TEG dehydration units are more likely located in
remote areas, we have differentiated between TEG dehydration units
located in densely populated areas (i.e., those located within the UA
plus offset and UC boundary) and those located outside those areas. 

Comment:  One commenter (0028) said that the definition of urban area
should be based on the most recent census data.  As the census data are
updated and sources change from a rural to an urban classification,
those sources should be considered existing sources.  To classify them
as new would be confusing to small businesses.  

Response:  Under the final rule, the definitions of UA and UC are based
on the most recent data from the Bureau of Census, which currently are
the 2000 Census data.  When census data are updated, we will propose to
amend this rule to reflect the new data if necessary.  

Exemptions

Define Low-Risk Subcategory

Comment:  Several commenters (0023/0024, 0032/0033, 0034) asked that EPA
consider creating a source category that can be exempted from the
regulations if the facilities can demonstrate insignificant health risk.
 The commenters (0023/0024, 0032/0033, 0034, 0043) cited such "low-risk"
subcategories in other MACT rules such as the Plywood and Composite Wood
Products MACT (40 CFR 63, subpart DDDD).  They believe that such an
exemption would be especially important if EPA adopts either of the
geographical applicability options in the proposed rule.  Two commenters
(0032/0033) indicated that an appendix is also needed for the proposed
rule that identifies the methodology and criteria for demonstrating that
an affected source is part of the low-risk subcategory – analogous to
Appendix B for subpart DDDD.  One of the commenters (0034) suggested
that these provisions include a simple look-up chart of criteria for
exclusion based on risk.

Response:  We interpret the commenters' primarily concerns to be the
regulation of TEG dehydration units in truly remote locations due to the
broad applicability that a county-based approach and a national approach
would require.  We believe that we have addressed the commenters'
concerns by requiring use of control devices only on sources located
within UA plus offset and UC boundaries and management practices (i.e.,
optimized glycol circulation rate) for sources located outside of UA
plus offset and UC boundaries.  

Distance Exemption

	Comment:  Commenters (0032/0033, 0034) said if the U-1/U-2 approach is
retained, provisions should be added that allow an affected unit to be
excluded based, for example, upon the filing of a certification and map
showing that the unit is not located within any municipality boundaries
or incorporated area, and is geographically distant from occupied
residences; or based upon the approaches suggested by INGAA and used in
other EPA rules, which take low risk into account based upon a simple
look-up chart of criteria for exclusion.  Such approaches will allow
sources in rural areas which are not contributing to any problem to be
excluded easily without the need for site-specific demonstrations or
modeling or the use of expensive consulting services.

	One commenter (0043) suggested a risk-based approach that looks at
emissions and the distance to the nearest receptor.  They supported this
position by stating that the area source rules are meant to protect
human health, therefore controlling offsite impacts where human health
could actually be affected is the most sensible option.

Several commenters (0022, 0026, 0030, 0035/0036) requested that EPA
consider (and adopt) a risk-distance calculation to impose controls only
on those area sources that could be seen to present an unacceptable risk
of exposure to nearby receptors.  Specifically, these commenters
suggested that, instead of applying area source controls nationwide or
in rural areas of counties that contain isolated urbanized areas, EPA
adopt a risk-distance approach calculation outlined in API publication,
API 4644, "A Methodology for Estimating Incremental Benzene Exposures
and Risks Associated with Glycol Dehydrators."  A copy of this
publication was provided (0036).

The commenters (0022, 0026, 0030, 0035/0036) pointed out that there is a
history of discussions between the industry and EPA on this approach. 
Commenters recommended that EPA adopt this screening method, which uses
the PC-based "SimRisk" model, during discussions surrounding the 1998
proposal, and EPA raised objections to using this method due to its
complexity.  The commenters indicated a willingness to address the
perceived problem and to make implementation of this risk-distance
method more user-friendly.  The commenters also pointed out that EPA
also objected in 1998 to the risk-distance calculation because the
method focused solely on protection of the most exposed individual
rather than the general population.  The commenters said that this
objection contradicts EPA's own risk analyses, which generally use risk
to the most exposed individual as a key measure in risk assessment.

Further, the commenters cited various examples of other regulations and
programs where similar approaches are utilized.  These include the
NESHAP for Industrial, Commercial, and Institutional Boilers and Process
Heaters, the NESHAP for Plywood and Composite Wood Products, the
California Air Toxic Hot Spots program, and the California Air Resources
Board land-use handbook.

Response:  As we stated in the previous response, we do not believe that
we have authority to exempt from regulation sources that have been
identified as necessary to reach the statutory 90% target.  However, we
believe that we have addressed the commenters’ concern by requiring
control devices on sources in densely populated areas.

Size Cutoffs

Comment:  Four commenters (0025, 0026, 0028, 0034) supported the
proposed provisions that exempt sources with gas throughputs less than
85,000 standard cubic meters per day (3 million cubic feet per day) or
actual average emissions of benzene from the TEG dehydration unit
process vent to the atmosphere less than 0.9 Mg/yr (1 ton/year).  One
of the commenters (0028) indicated that they think it is a good idea to
base applicability on flow rate, as it is easy to measure and is already
done for sales purpose.  They assumed that the 3 million cubic feet per
day gas flow rate was equivalent to the 1 ton/year benzene emissions
level, and pointed out that stack testing to determine the emissions
rate would be much more expensive than flow rate measurement.  One of
the commenters (0025) stated their belief than any lower cutoff
threshold would not be authorized by law, based on the fact the Agency
has not demonstrated either that controls exist for these types of
sources or that any such controls would be cost-effective.

Response:  We appreciate the commenters' support regarding the
Agency’s decision not to impose control equipment requirements on
sources with flow rate or benzene emissions below the cutoffs.  The
commenter is incorrect in assuming that the 3-MMscf/day cutoff is
equivalent to the 1-tpy benzene emissions cutoff.  These cutoffs are the
points below which we have not found control equipment being used and
have also determined that it would not be cost effective to control.  It
should be noted that sources may determine emissions using the GLYCalcTM
program in lieu of emissions testing.

Comment:  While they supported the concept, one commenter (0027) stated
their belief that the 3 million cubic feet per day gas throughput
exemption level was too low.  They wrote that, because of the typical
upstream aggregation of streams in the Appalachian Basin, these levels
will make many marginal streams subject.  As evidence, they cited that
the average production for an Ohio well is only 5 million cubic feet per
day.  To guard against making these small aggregated sources subject to
the rule, the commenter (0027) suggested that this exemption level be
changed to 5 million cubic feet per day.

Response:  We evaluated the difference between controlling only TEG
dehydration units with natural gas throughputs greater than 5 MMscf/day
and those with throughputs greater than 3 MMscf/day.  We determined
that the cost effectiveness associated with the 3 MMscf/day option
($2,400/ton) was reasonable.  Further, the commenter did not provide us
with additional information showing no TEG units below 5 MMscf/day are
controlled.  Therefore, the final rule retains the 3-MMscf/day
throughput exemption.

	Comment:  Another commenter (0030) supported the retention of the
proposed exemptions based on facility-level throughput thresholds in the
current rule that exempt facilities with annual average natural gas
throughput of 18,400 standard cubic meters per day (650 cubic feet per
day) or hydrocarbon liquid throughput of 39,700 liters per day
(250 barrels per day).  The commenter (0030) stated that this exemption
allows small production facilities that are incremental contributors to
HAP emissions to easily understand whether they need to further assess
their operations to determine whether an associated TEG dehydrator has
an actual annual average flow rate of natural gas exceeding 85,000
standard cubic meters per day or actual average benzene emissions of
more than one ton/yr.

	Response:  The intent of the exemption in §63.760(e)(2) was to provide
a cutoff so that very small facilities would not have to perform an
evaluation to determine whether or not they were major sources.
Therefore, section only applies to major sources and we have clarified
this in the final rule.  (Note:  the exemption in 760(e)(2) requires
that natural gas throughput be below 18,400 standard cubic meters per
day and 39,700 liters per day, rather than or as stated by the
commenter.

Decline Rates

	Comment:  Commenters (0022, 0026, 0030, 0031, 0035/0036, 0043) said EPA
should allow the facility operator to establish a true measure of
emissions for area sources prior to installation of controls.  Such a
provision would recognize that production in oil and gas operations
typically declines significantly over a short period of time after
initial startup.  Some commenters (0031, 0043) said that for dehydrators
with a design capacity of 10 MMscf/day or less that the operator should
be allowed 1 year to establish actual annual average flow rates to
determine if controls under the area source rule are required. 
Commenter 0031 added that the use of actual throughput volumes is
recommended because it is easy to determine.  This is a factor at small
independent operators who lack the personnel or expertise to conduct
complex emission calculations.

	Commenters (0022, 0026, 0030, 0035/0036) said determining whether a new
source is an affected source should be based on a known decline curve in
a producing field.  The commenters noted the NESHAP currently applicable
to major sources in the E&P industry allows facility operators to take
the decline in production into account in determining potential to emit
(PTE) for certain existing facilities.  EPA should extend this
philosophy to new wells by allowing PTE and exemption calculations to
incorporate representative historical data from existing wells within
the same field as new wells.  Specifically, EPA should add a new
subparagraph (iv) to §63.760(a)(1) to allow an operator placing a new
well online to make the determination using the daily average production
rate for the first year of operation of other representative wells
within the same field as the new well.  The rule would incorporate the
same conservative multiplier (1.2) that applies in §63.760(a)(1)(i)(A)
to existing wells.  To qualify, the TEG dehydrator would have to be
installed within an oil and gas production field with a demonstrated
history of production decline.  If the facility operator can
demonstrate, to the satisfaction of the Administrator, that the TEG
dehydrator is reasonably likely to meet the exemption levels in Subpart
HH during the first year of operations (i.e., annual average flowrate of
less than 3 million standard cubic feet per day (85 thousand cubic
meters per day) or actual average emissions of benzene of less than one
ton per year (0.90 megagrams per year)), based on that demonstrated
history of production decline and the first year production rates of the
other representative wells in the field, the new TEG dehydrator would
not be an affected major source or area source .

	Response:   New sources are required under section 112(i) of the CAA
and the General Provisions [40 CFR 63.6(b)] to be in compliance with
applicable regulations upon startup or upon the effective date of the
regulation (whichever is later).  Therefore, it is not appropriate to
include a provision in the final rule that would provide new sources an
extension of the compliance date.  Further, we believe that new sources
have options at their disposal to take the decline in production into
account.  For example, since the 3-MMscf/day cutoff is an annual
average, not an instantaneous value, once it becomes apparent that the
dehydration unit's throughput could exceed the 3-MMscf/day cutoff, a
source has the option to regulate the flow to either maintain a
throughput less than 3 MMscf/day or to install controls.  Thus, the
final rule does not contain a provision for incorporating decline rates.

Transition Policy On "Potential To Emit"

	Comment:  Commenters (0022, 0026, 0030, 0035/0036) asked EPA to verify
the continued effect of the Potential to Emit transition policy until
the completion of the PTE rulemaking process.  The commenters stated
that they understand that the provisions of the transition policy that
recognize state-enforceable limits are still in effect and that EPA
personnel have indicated that EPA intends to revive its pursuit of
resolution on this issue.  The commenters were particularly interested
in the continuing ability of a source to rely on a "practically
enforceable, state-enforceable limit" to restrict PTE.  Commenter 0022
said this is important for the great number of small oil and gas
dehydration units that will clearly be able to demonstrate actual
compliance with the proposed flow rate or annual benzene emissions
cutoff limits by being "subject to state-enforceable limits that are
enforceable as a practical matter."

	Response:  The PTE transition policy expired on December 31, 2000, with
a six-month extension for air permitting agencies that demonstrated the
need for the extension.  However, the state-enforceable provision of the
transition policy, which allows a source to rely on a practically
enforceable, state-enforceable limit to restrict its PTE, will remain in
effect until we have completed rulemaking on the term “potential to
emit.”

Documentation to Prove Control Equipment Exemptions

	Comment:  One commenter (0031) said EPA should minimize documentation
for sources to prove they are not subject to the requirements in the
final rule.  Emission calculations on all well sites would be costly and
burdensome.  Instead, basic data such as actual annual average flow rate
of natural gas should be sufficient documentation.

	Response:  The proposed rule does not require emission calculations on
all well sites.  First, only well sites that have a TEG dehydration unit
onsite are subject to the emission reduction requirements in the area
source standard.  Second, to qualify for the exemption, benzene emission
calculations are only necessary for TEG dehydration units with an actual
annual throughput greater than or equal to 3 MMscf/day.  Any TEG
dehydration unit with an actual annual throughput less than 3 MMscf/day
is exempt from emission reduction requirements.  Therefore, we believe
that our documentation requirements are sufficiently minimized and we
have not made any changes to the final rule in response to this comment.

Control Requirements

Controls for Remote/Unmanned Sources

	Comment:  Commenters (0022, 0023/0024, 0026, 0030, 0032/0033, 0034,
0035/0036) said if EPA imposes controls on TEG dehydrators outside of
Urban-1 areas, it should adopt a separate (lesser) control standard for
those remote area sources for the following reasons:

It is not justified based on health effects

Practical considerations prevent operators from achieving the 95%
control efficiency on remote, unmanned TEG dehydrators.

	Commenters (0022, 0026, 0030, 0035/0036) said that in order to meet the
95% control efficiency or the outlet concentration, an operator
generally has to install a system with a forced draft fan for the
condenser and a flare or vapor recovery system.  Many remote sources do
not have an electric power supply, which precludes using a forced draft
fan.  Routing the vapors to the firebox or fire-tube is not practical in
all situations because the high water vapor content can extinguish the
fire.  While flares and vapor recovery systems address this problem,
they require frequent monitoring, which is a problem at unmanned sites
that are only visited infrequently.  The lack of electric power supply
would make certain automated monitoring systems impossible.

	Commenters (0022, 0026, 0030, 0035/0036) said EPA should adopt a
separate GACT standard for facilities outside of "Urban-1" areas and
"urbanized areas."  The 95-percent control efficiency standard could
still apply in Urban-1 areas and "urbanized areas," but it would not
otherwise apply to area source TEG dehydrators.  The commenters
recommended that EPA set GACT for facilities that are not located in
Urban-1 or urbanized areas as a reduction of benzene to a level of less
than 1 tpy, and remove the 95-percent control efficiency requirement. 
Commenter 0022 added that GACT could also be considered as the
installation of a flash tank/condenser or incinerator process.

	Response:  We agree with the commenters that it is reasonable to
require a higher level of emission reductions for TEG dehydration units
located in more densely populated areas.  We also recognize that the oil
and natural gas source category is unique because there are many area
sources that are located in remote or rural areas.  For these reasons
and the reasons discussed above, we have subcategorized to differentiate
between those sources above the cutoff levels identified above that are
located inside UA plus offset and UC boundaries and those located
outside such boundaries.  We require installation of control equipment
for TEG dehydration units located inside UA plus offset and UC
boundaries and management practices (i.e., optimized glycol circulation
rate) for units located outside UA plus offset and UC boundaries.  We
believe that this approach addresses the commenters’ concerns
regarding the control of remote or rural facilities.

	Comment:  Commenters (0032/0033, 0034) said EPA should conduct
additional analysis associated with the burden and cost imposed, and
also the potential for unique technical issues associated with control
effectiveness, for implementing a standard at remote sites.

	Response:  Because we are not requiring add-on controls at sources
located outside a UA plus offset and UC boundary, it is not necessary to
revise the costs for the installation of add-on controls to address
burden at remote sites.  However, we have estimated impacts associated
with applying management practices (i.e., optimizing the TEG circulation
rate) for sources located outside of UA plus offset and UC boundaries. 
In order to be conservative in our analysis, we assumed that half of the
area sources located outside of UA plus offset and UC boundaries would
need to replace their glycol pump to achieve the optimum circulation
rate.   Because operating at the optimized TEG circulation rate results
in a reduction of natural gas losses, the annual cost includes a partial
cost savings, which reduces the annualized cost for glycol pump
replacement.  We estimate an annual cost of approximately $900 per area
source.

	Comment:  Some commenters (0031, 0043) said the 95-percent control
efficiency appears excessive for area sources, especially those that are
remote and unmanned.  A control efficiency of 80 percent was suggested. 
Commenter 0043 said achieving a 95-percent removal will require
combustion after a condenser, and there are safety issues associated
with post combustion.  Commenter 0031 said EPA has not justified the
95-percent efficiency rate, and EPA should provide justification for any
efficiency rate it proposes.

	Response:  As we stated in the preamble to the February 1998 proposed
area source standard (63 FR 6299), condensers and flares installed on
actual area source TEG dehydration units have been observed and we
believe that condensers capable of achieving 95 percent emission
reduction are technically feasible for area sources.  The commenters did
not provide any data showing their basis of the 80-percent control
efficiency.  Therefore, we have not changed the level of control for the
area source standards.  However, as stated previously, the final rule
distinguishes between sources located in densely populated areas and
those located in rural or remote areas.  TEG dehydration units with an
actual annual average throughput of 3 MMscf/day or more and benzene
emissions of 1 tpy or more, that are located outside a UA plus offset
and UC boundary, are required to implement management practices.

	Comment:  Commenter (0031) asked EPA to clarify if a flare is required
to meet the efficiency rate, would it have to meet the flare standards
in 40 CFR 63, subpart A.

	Response:  The commenter is correct, flares are required to meet the
standards in the General Provisions, 40 CFR 63, subpart A [see
§63.771(d)(i)(iii)].

Start-up, Shutdown, Malfunction Plans

	Comment:  One commenter (0025) supported the decision to not require
submittal of malfunction reports as proposed in 1998 and prefers the
current proposal's requirements.  This would have been burdensome and
impractical, especially in remote locations that do not have full-time
operators onsite.

	Response:  We appreciate the commenter's support of the current
proposal's startup, shutdown, and malfunction report requirements.  No
changes to these requirements have been made to the final rule.

Reporting and Recordkeeping

	Comment:  Commenters (0022,0023/0024, 0026, 0030, 0032/0033, 0034,
0035/0036) said setting a GACT standard for remote area source TEG
dehydrators that eliminates the 95% control efficiency standard should
not require the same level of monitoring, recordkeeping, and reporting
that applies to major sources in truly urban areas.  Commenters (0022,
0026, 0030, 0035/0036) said a remote area source TEG should be required
to register the source with the agency implementing the control
requirements and demonstrate compliance with the GLYCalcTM calculation. 
The operator could be required to reconfirm that the source meets the
1-tpy standard through quarterly measurements.  

	Commenters (0023/0024, 0025, 0032/0033, 0034) added that compliance
with the startup, shutdown, and malfunction (SSM) requirements will be
difficult for remote, unmanned locations because these requirements
normally call for immediate action when there is a failure.  Commenters
(0032/0033, 0034) added that EPA has not taken into account previous
comments on SSM requirements that logging of events and submittal of
reports (including time constraints associated with immediate reporting
in some instances) impose issues that are difficult to address for
remote sites.

	Some commenters (0029, 0031) felt that EPA should simplify the
reporting and recordkeeping for area sources in general because many of
the units covered may be in remote areas and at unmanned facilities. 
Commenter 0031 added that smaller operators will be greatly impacted if
this level of detail is required.

	Response:  We believe that the recordkeeping and reporting requirements
for remote sources under the final rule are simple and reasonable. 
First, SSM procedures only apply to sources required to install add-on
controls, which are only required for sources located within the UA plus
offset and UC boundary.  Further, only sources with TEG dehydration
units with annual average throughput of 3 MMscf/day or more and benzene
emissions of 1 tpy or more are required to submit documentation
providing their location.  Sources with TEG dehydration units with
throughputs less than 3 MMscf/day or benzene emissions less than 1 tpy
are required to maintain records of the determination of these criteria
[§§63.764(e)(1) and 63.774(d)(1)]  but are not subject to any
reporting requirement.  

	Comment:  One commenter (0021) pointed out that the provisions
requiring that records be kept on site or accessible within 2 hours of
the dehydrator are impossible for the Rocky Mountain region, where
winter prevents access to the well site within 2 hours.

	Response:  Section 63.774(b)(1)(ii) states that records can be stored
in a central location either on computer or other means that provides
access within 2 hours after the request.  Therefore, records for
facilities located in regions that may be inaccessible may be stored at
an off-site location provided they are accessible within the 2-hour time
period.

Test Methods

	Comment:  Commenters (0032/0033, 0034) endorsed the addition of ASTM
D6420-99(2004) as an alternative test method to EPA Method 18.

	Response:  We appreciate the commenters' endorsement of this
alternative test method.	 

Impacts Analyses

Several commenters (0021, 0027, 0031, 0034) referred to several aspects
of EPA's economic impact analysis, including: the TEG dehydrator
population estimate, the cost impact analysis, and the impact of the
rule on regulated sources.  One commenter (0021) stated that EPA's
economic impact analysis is based on assumptions that are not "supported
by the proposed rule, references, discussion, or logic."  The commenter
(0021) recommended that EPA should prepare an accurate estimate of the
cost of the rule.

TEG Dehydrator Population Estimate

	Comment:  Two commenters (0021, 0031) commented on EPA's estimate of
the TEG dehydrator population.  One commenter (0021) did not agree with
EPA's approach for estimating the number of wells and stated that the
number appeared low.  The commenter's (0021) arguments had three main
points.  First, the commenter (0021) noted that EPA relied on a 1996
discussion with representatives of API for the number of wells having
specified gas production ranges.  According to the commenter (0021), API
represents major operators, who typically find small gas sources
uneconomical, meaning that the estimates may be biased.  The commenter
(0021) stated that discussions with representatives of Railroad
Commission of Texas show that the State of Texas does not routinely
record statistics comparing the number of gas wells as a function of
production rate.  The commenter (0021) stated that these statistics were
available for a fee and suggested that EPA should have contracted for
them.  In addition, the commenter (0021) stated that they did not
believe API worked with 33 State Agencies to prepare the data.

	Second, the commenter (0021) referred to the fact that area source
statistics were derived from the number of Gas Processing Plants
estimated to exist in 2003.   The commenter (0021) maintained that the
correlation between gas plants and TEG area sources was unsupported. 
The commenter (0021) stated that according to the Railroad Commission of
Texas there was no data to support the assumption.  

Third, the commenter (0021) stated that EPA provided no insight into how
the TEG/Gas Plant correlation was derived.  Specifically, the commenter
(0021) stated that the table of Area Source TEG Dehydration Units by
throughput from 0.1 to 5 MMscf/day was not supported.  According to the
commenter (0021), EPA's estimate of the number of dehydration units is
low.  In support of their argument, the commenter (0021) referred to
personal communications with the Railroad Commission of Texas engineers,
which indicated that EPA's estimate of the number of TEG dehydration
units is about the same as the number Texas record show for all E&P
dehydrators in Texas in the range from 0.1 to 5 MMscf/day.

The commenter (0021) also referred to EPA's assumption in the current
estimate that 10 percent of the TEG dehydrators would use flash tanks. 
According to the commenter (0021), this assumption was based on the
original proposal background information document where EPA did not
recommend the use of the 10-percent flash tank estimate, "but simply
suggests if 10 percent is used, the results are...."  The commenter
(0021) maintained that these statements are misrepresentations of the
facts and cast doubts on the rulemaking process.

The commenter (0021) stated that they have attempted to obtain TEG
dehydration unit population without success.  According to the commenter
(0021), the Independent Petroleum Producers Association (IPPA) do not
collect these data.  Further, the commenter (0021) stated that
representatives of the Railroad Commission of Texas have demonstrated
that they do not have a means of distinguishing TEG dehydration units,
regardless of throughput capacities.

The commenter (0021) supported EPA's emission estimates based on
GRI-GLYCalc, but stated that using these emission estimates with
unsupported population estimates appears questionable.

One commenter (0031) questioned where EPA's impact estimate came from. 
The commenter (0031) clarified that they did not understand how EPA
estimated that 2,200 sources would be impacted by applicability Option 1
(national applicability) and 1,050 sources would be impacted by Option 2
(Urban-1/Urban-2 applicability).  The commenter stated that EPA needed
to justify its data and assumption.

	One commenter (0021) stated that EPA's estimates of TEG dehydrator
population are questionable, and therefore any cost/benefit analysis
based on these population estimates is in question.

	Response:   According to the Railroad Commission of Texas, they do not
maintain information about the glycol dehydration unit population in the
State of Texas.  The Commission does maintain data on the number of
active wells in the State. We believe that the commenter may have
confused the number of wells with the number of dehydration units.

	Our estimate of the population of TEG dehydration units was based on a
procedure developed by a consultant to the American Petroleum Institute
(API) and reasonable apportioning of the number of TEG dehydration units
with actual natural gas throughputs ranging from 0.1 to 5 MMscf/day. 
Throughput sizes were apportioned across the size ranges based on the
number of wells, the number of processing plants and the volume of gas
produced.    We used this estimate to compare control options.  No
additional information regarding the population of TEG dehydration units
was provided by the commenters.  Therefore, we have not adjusted our
estimates of the TEG dehydration unit population.

No New Area Source Assumption

	Comment:  Commenter (0043) disagreed with EPA's assumption that any new
sources would be major sources.  The commenter said the opposite is true
because industry tries to avoid the title V process by limiting
emissions below major source thresholds.  Therefore, the impact of the
proposal is much greater than assumed.

	Commenter (0031) questioned the new source assumption and asked for
clarification regarding the resulting title V impacts.  The commenter
was concerned that all sources subject to GACT would be required to
obtain title V permits, which would be burdensome and costly.

	Response:  In response to this comment, we reevaluated our assumptions
related to the number of new sources that would be subject to the area
source rule.  We estimate that there will be 423 new area sources
constructed in the first three years after promulgation of the final
rule, based on the trend in the number of new wells drilled from 2002 to
2005 that we used to estimate future well drilling activity.  Of these
423 new sources, we estimate that 6 sources will be located within UA
plus offset and UC boundaries.  We believe that the resulting costs and
burden of complying with the area source standards for these sources are
reasonable.  Area sources are not required to get Title V permits under
subpart HH.

Unmanned Facilities Use of Flares

	Comment:  Commenters (0032/0033, 0034) said many remote, unmanned
sources in West Virginia use flares rather than condensers due to gas
characteristics in the area.  Therefore, controls here may prove to be
more costly than assumed by EPA for sources in remote areas that are
actually having less impact than for most other areas.

	Response:  We believe that the final rule, which only requires
management practices for sources outside of UA plus offset and UC
boundaries, addresses the commenters' concerns regarding the costs of
add-on controls (i.e., flares) for remote sources.	

Sensitivity To Operating Costs

	Comment:  Commenter 0027 was concerned that the increased regulatory
costs associated with the proposal will result in the premature
abandonment and permanent loss of a significant source of U.S. natural
gas supplies.  The vast majority of production in the Appalachian Basin
states is economically marginal, and these natural gas streams are
extremely sensitive to any increases in operating costs, including
increased regulatory costs.  These streams are often aggregated from
hundreds and thousands of wells behind a TEG dehydration unit.  Applying
standards to the TEG dehydration unit will push the regulatory costs
upstream, resulting in the premature loss of otherwise viable domestic
marginal natural gas supplies.

	One commenter (0031) said small independent operators, which are
prevalent in the industry, do not have the personnel or the expertise to
evaluate or implement the proposed rules if they become final.  Smaller
operators will have to hire consultants to address the additional
requirements, which create an additional cost burden.  EPA needs to
reconsider the impacts to smaller operators.

	Response:  As a part of the rulemaking process, we are required to
evaluate the impact a rule may have on small businesses.  Our evaluation
was based on the maximum level of control required at a single source,
and the associated costs incurred, in comparison to the minimum revenue
realized from that source.  This analysis is presented in section V.C
(Regulatory Flexibility Act) of the preamble to the July 8, 2005
supplemental proposal (70 FR 39449).  We estimated that the proposed
area source standards would have costs significantly less than
1 percent of revenues.  Based on this estimate, we determined that the
annual cost of control for facilities affected by the proposed rule is
not sufficient to generate a significant impact on a substantial number
of small entities.

Comments Received on February 6, 1998, Proposed Rule

	In addition to the public comments received on the July 8, 2005,
supplemental proposal, we also received comments on the original
February 6, 1998, proposal related to the area source standards.  These
comments were not addressed since final action with respect to area
sources was deferred.  This section provides a summary of the comments
received on the February 6, 1998 proposed area source standards and our
responses to these comments.  The legacy docket for these comments is
Docket No. A-94-04.  Table 2 presents a listing of only the persons
submitting written comments on the area source standards (i.e.,
commenters making comments on the major source rule are not shown),
their affiliations, and the docket item number for their comment. Table
2.  List of Commenters on Proposed Standards for Oil and Natural Gas
Production Industrya

Docket Item Numberb	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-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-10	T. Hutchins

El Paso Energy Company

V-D-11	R. Metcalf

Louisiana Mid-Continent Oil and Gas Association

801 North Boulevard, Suite 201

Baton Rouge, Louisiana  70802

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-19	W. Airey

Vorys, Sater, Seymour, and Pease LLP

52 East Gay Street

P.O. Box 1008

Columbus, Ohio  43216

IV-D-20	K. Beckett

Jackson & Kelly

1600 Laidley Tower

P.O. Box 553

Charleston, West Virginia  25322

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-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-29	M. Chytilo

Environmental Defense Center

906 Garden Street

Santa Barbara, California  93101

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-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-38	M. Fish

Enron Oil & Gas Company

P.O. Box 4362

Houston, Texas  77210

IV-G-02	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-9	P. Bennett

KN Energy Inc.

One Allen Center

500 Dallas Street, Suite 500

Houston, Texas  77002

IV-G-12	M. Fox

New Century Energies

P.O. Box 840

Denver, Colorado  80202

a Only those commenters that submitted comments on the area source
standard are included in this table.

 ADVANCE \u 2 b ADVANCE \d 2  The docket number for this project is
A-94-04.  Dockets are on file at EPA Headquarters in Washington, D.C.

Definition of Urban Area

	Several commenters responded to EPA’s request for comments on EPA’s
proposed definition of urban area.  The major comments received about
the definition of urban area were on the following issues: (1) the
interpretation of the language in section 112(n)(4); (2) the
distinction between “urban” and  “urbanized”; (3) the
classification of entire counties as urban; (4) the classification of
counties as urban that are truly rural, and (5) the inclusion of an
option for a “risk-distance” approach.

Interpretation of Section 112(n)(4)

 	Comment:  Nine commenters (IV-D-04, IV-D-08, IV-D-15, IV-D-19,
IV-D-20, IV-D-22, IV-D-26, IV-D-34, and IV-D-38) stated that EPA’s
definition of urban area was inconsistent with the statutory language in
section 112(n)(4)(B).  The commenters stated that the language in
section 112(n)(4)(B) prohibits EPA from listing “oil and gas
production wells (with its associated equipment)” as area sources
except those located in a MSA with a population of one million or more. 
The commenters stated that EPA’s proposed definition is too broad
because it includes county-wide MSAs with a population of more than
250,000, Census-defined “urbanized areas,” and apparently
Census-defined small-town “urban areas” with a population of at
least 2,500.  One commenter (IV-D-15) noted the proposed definition is
unclear on the status of small-town “urban areas,” but the preamble
(63 FR 6293 through 6294) and docket materials suggest that EPA intends
to include them.  Another commenter (IV-D-08) recommended that EPA make
the following revision to the area source location applicability
criteria such that:

...only TEG area sources located within metropolitan statistical areas
or consolidated metropolitan statistical areas with a population in
excess of 1 million are subject to the finding.

The commenter stated that this revision reflects Congress’ intent in
section 112(n)(4) and focuses on congested areas.  

	One commenter (IV-D-38) also requested that a modification to
§63.764(e) be made if EPA finds the risk of adverse effects to public
health to be more than negligible.  According to the commenter, Congress
intended that glycol dehydration units should be included as associated
equipment.  Therefore, the commenter recommended the following change:

(e)  . . .  In addition, the owner or operator is exempt from the
requirements of paragraph (d) of this section if the glycol dehydration
unit is not located in a metropolitan statistical area (MSA) with a
population in excess of 1 million.

The commenter also stated that the definition of urban area in the
proposed regulation is unnecessary and should be eliminated.

	Response:  As we indicated in section 2.3.4 of this document
(Geographic Applicability Criteria), the provisions in section
112(n)(4)(B) of the Clean Air Act as Amended in 1990 apply to "oil and
gas production wells (with its associated equipment)."  Based on our
interpretation of associated equipment, glycol dehydration units are not
considered as part of a well and its associated equipment.  The final
rule does not incorporate the changes suggested by the commenters.

Distinction between "Urban" and "Urbanized"

	Comment:  Five commenters (IV-D-08, IV-D-15, IV-D-22, IV-D-23, and
IV-D-24) stated that in its definition of Urban-2 counties, EPA had
confused the terms "urban" and "urbanized."  According to the
commenters, the expression "...areas that compromise one or more central
places and the adjacent densely settled surrounding fringe that together
have a minimum of 50,000 persons... ” appears to apply to urbanized
areas, yet the definition of Urban-2 states that the counties are
defined as " . . . all other counties designated as urban by the Bureau
of Census...." Two commenters (IV-D-08 and IV-D-24) stated that if EPA
intended to use the term "urbanized" rather than "urban," then the
determination of which counties are Urban-2 areas is incorrect.  One of
the commenters (IV-D-08) noted that the counties included in Figure 1 of
the preamble include both urban and urbanized areas.  According to two
commenters (IV-D-15 and IV-D-22), the preamble and docket materials
(Item A-94-04, II-I-9) suggest that EPA may apply area source controls
in areas that do not qualify as "urban" under Census classifications. 
The commenters stated that the preamble and docket materials have a
criterion that more than 50 percent of the county population must be
considered urban in determining which areas are Urban-2 areas.  The
commenters, along with another commenter (IV-D-23), recommended that EPA
modify the proposal to limit applicability to large metropolitan
statistical areas, "Urban-1" areas, and Census-designated "urbanized
areas."  The commenters stated that this modification would conform to
Congressional intent to limit area source controls to truly urban areas.
 One commenter (IV-D-22) stated that the Bureau of Census definition of
"urbanized areas" is more consistent with a risk finding for the source
category:

The Census Bureau delineates urbanized areas to provide a better
separation of urban and rural territory, population, and housing in the
vicinity of large places.  An urbanized area comprises one or more
places ("central place") and the adjacent densely settled surrounding
territory ("urban fringe") that together have a minimum of 50,000
persons.  The urban fringe generally consists of contiguous territory
having a density of at least 1,000 persons per square mile.

	One commenter (IV-D-24) suggested updating the Urban-2 area definition
as follows:

Urban-2 areas which are defined as all other areas designated as
urbanized areas by the Bureau of Census (areas which comprise one or
more central places and the densely settled surrounding fringe that
together have a minimum of 50,000 persons).  The urban fringe consists
of contiguous territory having a density of at least 1,000 persons per
square mile.

	Response:  The commenters were correct that there was an error in the
definition of Urban-2 in the preamble of the February 1998 proposed area
source standards.  However, the final rule does not utilize the
definition of Urban-2.  Instead, the final rule refers only to sources
that were located in Urban-1 counties and those that are not located in
Urban-1 counties (which includes counties that met the February 1998
definition of Urban-2 and rural).  This distinction is important when
determining new source status and is discussed in more detail in section
2.2 of this document (Compliance Date).

Classification of Entire Counties as Urban  

	Comment:  Six commenters (IV-D-08, IV-D-15, IV-D-20, IV-D-26, IV-D-27,
and IV-D-31) objected to designating entire counties as urban.  The
commenters stated that several counties in the U.S. that have large
geographic areas (e.g., Texas has counties with more than 1,000 square
miles) where most of the county area is rural with a small portion that
is a small city or metropolitan area.  These commenters were concerned
that the definition of an entire county as "urban" based on one small
population center would transform vast rural areas into urban areas. 
The commenters were also concerned that one urbanized center in the
center of a sparsely populated county could cause the entire county to
be classified as "urban" under the proposed method.  One commenter
(IV-D-20) mentioned that emissions of remote sources may not affect
urban areas because the sources are within a different airshed, given
the terrain and distance from the MSA.

	 One commenter (IV-D-15) also stated that EPA misapplied Census
classifications.  According to the commenter, MSAs cover entire counties
but other Census-designated categories do not:  they are limited to
Census blocks.  The commenter stated that the more heavily populated a
metropolitan area is, the broader it’s economic influence, making
expansion of MSAs to the county line appropriate.  The same expanded
influence does not apply to the small Census-designated areas.

	Two commenters (IV-D-15 and IV-D-22) recommended, and one commenter
(IV-D-34) supported, the following modification to the definition of
urban area to conform to legislative intent:

Urban area for the purposes of the area source determination is defined
by use of the U.S. Department of Commerce’s Bureau of the Census
statistical data to classify all land area in the United States into one
of the three classifications, as follows:

i.	Urban-1 areas, which consist of metropolitan statistical areas (MSAs)
with a population greater than 250,000;

ii.	Urban-2 areas, which are defined as all other areas designated as
"urbanized" by the Bureau of Census (areas that comprise one or more
central places and the adjacent densely settled surrounding fringe that
together have a minimum of 50,000 persons.  The urban fringe consists of
contiguous territory having a density of at least 1,000 persons per
square mile); or

iii.	Rural areas, which are all areas that are not designated as Urban-1
or Urban-2.

For consistency, the commenters also requested that EPA make a
conforming amendment to § 63.764(e):

(e)  . . .  In addition, the owner or operator is exempt from the
requirements of paragraph (d) of this section if the glycol dehydration
unit is not located in a county classified as an Urban area as defined
in Section 63.761.

	

	One commenter (IV-D-27) stated that given the legislative history of
section 112 of the CAA, it is unlikely that Congress considers entire
counties as urban.  Therefore, the commenter, along with another
commenter (IV-D-31), recommended that EPA use the following definition
of urban area published by the Department of Commerce (Statistical
Abstract of the United States, page 4, 1995 ed., "Urban and Rural") as a
basis for defining urban areas:

Urban area means (1) a county in a metropolitan statistical area (MSA)
with a population greater than 250,000 and (2) those other urban areas
in which the urban population comprises all persons living in (a) places
of 2,500 or more inhabitants incorporated as cities, villages, boroughs
(except in Alaska and New York), and towns (except in the New England
States, New York, and Wisconsin), but excluding those persons living in
the rural portions of extended cities (places with low population
density in one or more large parts of their area); (b) census designated
places (previously termed unincorporated) of 2,500 or more inhabitants;
and (c) other territory, incorporated or unincorporated, included in
urbanized areas.  An urbanized area comprises one or more places and the
adjacent densely settled surrounding territory that together have a
minimum population of 50,000 persons.  In all definitions, the
population not classified as urban constitutes the rural population.

	Thirteen commenters (IV-D-04, IV-D-08, IV-D-10, IV-D-15, IV-D-20,
IV-D-22, IV-D-23, IV-D-24, IV-D-30, IV-G-02, IV-G-03, IV-G-09, and
IV-G-12) were concerned that the current definition of an Urban-2 area
would cause many sources located in areas that are truly rural to be
subject to the area source requirement.  The commenters recommended that
EPA define "urban" to avoid covering large portions of land that are
truly "rural" in nature (e.g., largely agricultural land use,
undeveloped land, etc.).  Three commenters (IV-D-04, IV-D-30, and
IV-G-03) requested that EPA modify the definition of urban area not to
include, for control purposes, equipment located in rural counties with
small urban places.  According to three commenters (IV-D-08, IV-D-20,
and IV-D-22), subjecting many area sources to control requirements is
unjustified by the risk posed and is counter to the stated focus of the
area source determination.  One commenter (IV-D-10) recommended deleting
all references to Urban-2 to avoid including sources that have no
potential of impacting urban populations. 

	Response:  As we stated in section 2.3.4 of this document (Geographic
Applicability Criteria), we rejected Option 2, which is a county-based
scope based on the definitions of Urban-1 and Urban-2.  The final rule
regulates TEG dehydration units on a national scope, but differentiates
the level of control between sources located within UA plus offset and
UC boundaries and those outside of these boundaries.  

Risk-distance Option 

	Comment:  Sixteen commenters (IV-D-04, IV-D-08, IV-D-11, IV-D-14,
IV-D-16, IV-D-20, IV-D-22, IV-D-24, IV-D-26, IV-D-30, IV-D-32, IV-D-34,
IV-G-02, IV-G-03, IV-G-09, and IV-G-12) stated that the gas throughput
and benzene emission criteria for TEG dehydrators [3 MMscf/day and 1
tpy, respectively] were appropriate for determining area source
applicability.  However, the commenters requested that EPA revise the
location criteria to include a "risk-distance" approach for determining
whether a facility is subject to MACT.  The commenters recommended that
the definition of urban areas be tailored to address the risk identified
- maximum individual risk in areas where persons reside close to TEG
dehydration units.  In general, the commenters suggested that EPA use
the urban area definition to define areas where TEG units are
potentially covered by the area source regulation, and allow the
operators the option of using a risk-based methodology, such as that
described in API Publication 4644, to demonstrate that they pose a
negligible risk to nearby residents.  The commenters stated that using
such a methodology if it is determined that the health risks are
negligible (e.g., an most exposed individual cancer risk of less than
1x10 ADVANCE \u 2 -6 ADVANCE \d 2 ), the source would not be subject to
control.  Two commenters (IV-D-22 and IV-D-34) requested that EPA
publish in the Federal Register, the risk-distance graphs and equations
from API Publication 4644, along with examples, to make the approach
easy to apply.  The commenters also recommended if such a methodology is
allowed, that the owner or operator should be required to maintain, and
have readily available, keep documentation of the analysis.  Two
commenters (IV-D-08 and IV-D-20) proposed a revision to subpart HH to
provide a location applicability such that:

"...TEG area sources otherwise subject to the standard (because of gas
throughput, benzene emission levels, and location) should not be subject
to additional standard requirements if the operator demonstrated that
the most exposed individual health risk is negligible."

	One commenter (IV-D-16) recommended that EPA consider the population
density around the immediate area in which the unit is located, in the
same manner implemented in design considerations for pipelines by the
Department of Transportation (DOT).

	One commenter (IV-D-24) recommended defining area sources as those
located in "populous areas" rather than "urban areas" to eliminate
confusing language from the definition of area source.  The commenter
recommended that "populous areas" consist of MSAs greater than 250,000
people (or some higher threshold) and "urbanized areas" (i.e., areas
that comprise one or more central places and the adjacent densely
settled surrounding fringe that together have a minimum of 50,000
persons).

	Commenter IV-D-08 recommended the following definition for "urban and
rural areas:" 

Urban and rural areas:  For purposes of applicability of the area source
requirements of this subpart, portions of every county/parish in the
United States that are classified as either "urban" or "rural," based on
certain U.S. Department of Commerce Bureau of the Census
classifications.  Urban areas are: (i) those entire counties, parishes,
or equivalent delineated census areas [referred to herein as "counties"]
that fall wholly or partially within a consolidated metropolitan
statistical areas (CMSA) with a population greater than 1,000,000; and
(ii) those specified portions of all other counties which are defined by
the Bureau of the Census as "urbanized areas," comprising one or more
central places, and the adjacent urban fringe of contiguous territory
having a density of at least 1,000 persons per square mile, that
together have a minimum of 50,000 persons.  Rural areas are all other
areas of the United States, which are not classified as urban as defined
above.

	Response:  As we stated in our response in section 2.4.2 of this
document (Distance Exemption), we do not believe that we have authority
to exempt from regulating sources that have been identified as necessary
to reach the statutory 90% target.  However, we believe that we have
addressed the commenters' concerns by requiring use of control device
for sources located within a UA plus offset and UC boundary and
management practices (i.e., optimized glycol circulation rate) for
sources located outside these boundaries.

	Comment:  Commenter IV-D-16 recommended that, in subpart HH, EPA
publish a list of all counties considered Urban and suggested that a
pointer to Table 3 Urban Counties as appropriate.  The commenter stated
that pointing to other groups’ lists is inappropriate since EPA has no
control or assurance that the other groups will not discontinue or
rename the document being referred to.   The commenter referred to the
Internal Revenue Service (IRS) Publication 534, which is pointed to by
the new source performance standard (NSPS) General Provision, which has
been discontinued and is impossible to obtain.

	Response:  We included a current listing of urban counties in the
publicly-available docket to this proposed rulemaking, docket item
II-I-9 of EPA Air Docket A-94-04.

Area Source Regulation

	Comment:  One commenter (IV-D-01) stated that if the EPA is to make an
urban versus non-urban designation for determining which area sources
are to be affected, it should be addressed under § 63.760
(Applicability).

	Response:  The final rules applies on a national scope, therefore no
provisions are necessary stating that area sources in rural areas are
not subject to subpart HH.

	Comment:  One commenter (IV-D-05) suggested that all references to TEG
units be removed from the area source requirements.  According to the
commenter, specifying TEG provides an exemption for other glycol units. 
The commenter stated that specifying a glycol type was redundant and
would create confusion and loopholes allowing units to use other types
of glycol to avoid applicability.  The commenter stated that the type of
glycol should not matter.  Another commenter (IV-D-35) stated that they
interpreted subpart HH to require area sources to use triethylene glycol
and questioned why the regulation does not address units that use
ethylene glycol. The commenter was also concerned that the use of
ethylene glycol would exempt certain area sources.

	Response: As we stated in the July 2005 preamble to the supplemental
proposal, the 1999 area source listing in the UATS was based on
emissions information showing that TEG dehydration units contributed
significantly to nationwide emissions of benzene from area sources in
urban areas.  Furthermore, TEG dehydration units account for
approximately 90 percent of the HAP emissions at oil and natural gas
production facilities.  Therefore, we focused on regulating benzene
emissions from TEG dehydration units, and the final rule regulates only
TEG dehydration units located at area sources.	

	Comment:  One commenter (IV-D-06) stated that subpart HHH should not
apply to area sources.  The commenter provided three reasons for not
regulating area sources under subpart HHH:

1.	Glycol dehydration units for natural gas transmission and storage are
typically located remote areas where there is little potential for
harmful levels of HAP exposure.  The commenter referred to section
112(c) of the CAA which allows EPA to regulate area sources only if
there is “a threat of adverse effects to human health or the
environment” warranting regulations.  The commenter also pointed to
section 112(k) which indicates that Congress was concerned about urban
area sources.  Therefore, the commenter stated that it would be unlikely
that regulation of area sources would be needed in subpart HHH.

2.	Emissions from glycol dehydration units are below the major source
levels, which would reduce the likelihood for harmful levels of HAP
exposure even if some of these units were located in more heavily
populated areas.

3.	Regulating all area sources, in order to capture the few that pose an
unreasonable risk would be unfair.  The commenter explained that a
“one size fits all” rule on every area source would be unjustified
in the unlikely event that a few area sources present an unreasonable
risk.  The commenter stated that the regulatory burdens would be for
only a limited subset of area sources.  The commenter recommended that
EPA should address this issue through the “residual risk” provisions
of section 112(f) of the Act rather than by setting an area source
standard under §§ 112(c) and (d). 

	Three commenters (IV-D-07, IV-D-31, and IV-G-09) requested that EPA not
regulate area sources in the transmission and storage source category
until more data could be collected to determine whether an area source
regulation is required.  One commenter (IV-D-07) stated that if these
sources are to be regulated, then one rulemaking addressing both major
and area sources is essential.  According to two of the commenters
(IV-D-07 and IV-D-31), transmission and storage operations tend to be
centrally located and are usually much larger in volume but much lower
in HAP emissions than production facilities.  The commenters stated
that, under proposed subpart HHH, very few facilities would qualify as
area sources.  One commenter (IV-D-07) provided an example of facilities
that accepted operating restrictions to qualify as synthetic minor
sources.  The commenter contended that the need for regulating these
facilities is eliminated by the need for operating permits.  The
commenter strongly recommended that, if EPA decides to include area
sources in subpart HHH, EPA should retain the option that area sources
be determined solely on the basis of the classification of the county in
which it is located.  One commenter (IV-D-31) stated that EPA is
considering an area source determination for natural gas transmission
and storage facilities without providing the industry the benefit of a
formalized information collection process and evaluation.  The commenter
offered to assist EPA in gathering data to confirm that such units do
not pose unacceptable risks warranting area source standards. 

	Although one commenter (IV-G-09) does not presently have data that
compares HAP emissions from the initial dehydration to subsequent
dehydration after pipeline quality gas is transported and temporarily
stored underground, the commenter maintained that, logically, HAP
emissions should be less from the subsequent dehydration.  The commenter
was concerned that there is not enough lead time for operators to gather
data requested by EPA on area source dehydration units and that random
submission of data may not be representative of the industry and could
result in errors in impact estimates.  The commenter suggested that EPA
work with pipeline operators and their trade associations to collect
information about area source dehydration in a uniform manner, and from
a statistically significant sample of the transmission industry. 

	Response:  We are not regulating TEG dehydration units located at
natural gas transmission and storage facilities under subpart HHH. 
These units were not included in the July 1999 listing under sections
112(c)(3) and 112(k)(3)(B) of the CAA.

	Comment:  One commenter (IV-G-09) agreed that control of HAP emissions
below stated cutoff levels is not cost effective for oil and gas
production area source dehydration units.

	Response:  We appreciate the commenter's support regarding the cutoffs
for area sources (i.e., TEG dehydration units with an actual average
annual natural gas throughput less than 3 MMscf/day or actual average
benzene emissions less than 1 tpy do not have any control requirements).

	Comment:  Three commenters (IV-D-22, IV-D-30, and IV-G-03) recommended,
and one commenter (IV-D-15) supported, that EPA remove the 95 percent
control efficiency from area sources, and adopt a revised GACT standard
for area source triethylene glycol dehydration units as an alternative
that would protect human health and the environment at lower cost.  One
commenter (IV-D-08) also requested that EPA revise the GACT standard to
eliminate noncompliance instances for area sources.  The commenters
recommended the following control strategy:

Define GACT as reduction of benzene emissions to a level less than 1 tpy
or as installation and operation of a flash tank and glycol dehydrator
(or equivalent control device) [Note:  The commenter mentioned “glycol
dehydrator” but it was assumed that they meant “condenser.”] for
area source units subject to the rule.  No control efficiency
calculation will be required.  An operator would register with the
implementing agency to demonstrate that the control equipment operates
within unit design standards.

	One commenter (IV-D-30) also recommended that the owner or operator
would be given further flexibility to monitor any operating parameter
appropriate to the unit’s characteristics in order to demonstrate
compliance.  Two commenters (IV-D-08 and IV-G-03) also recommended that
EPA implement the following to provide a cost-effective solution and
provide the desired level of environmental protection:

Require monitoring of the condenser exhaust temperature and the use of
GLYCalc or equivalent methodology (e.g., a design analysis) to determine
the minimum exhaust temperature that will result in a benzene emission
level less than 1 tpy.

Require monitoring to show that the condenser (or other control device)
is operating properly.  For condensers, the commenter proposed that
monitoring should consist of measuring the change in temperature (T)
across the condenser.  The minimum required T should be determined by
design analysis for each unit subject to this GACT standard.

Add risk-distance as another applicability criterion to the existing
criteria (i.e., 3 MMscf/day, 1 tpy benzene, urban-1, urban-2) to define
area source TEG units subject to GACT control requirements.  Addition of
this criterion will control HAP emissions in those situations where HAP
emissions could negatively impact offsite receptors.

	One commenter (IV-D-05) noted that there were no differences in the
requirements for glycol units located at an area source and glycol units
located at a major source.  Therefore, the commenter suggested that EPA
state the following:

“...glycol units that process 3 MMscf/day or more and that emit 1 tpy
of benzene and are not located in a Rural area are subject to the
control, recordkeeping, monitoring, etc.”

	

	Response:  In developing standards for area sources, we evaluated
alternative methods for regulating area source TEG dehydration units. 
Through a review of available information and with support from results
using GLYCalc, we believe that the selected control requirements are
technically and economically achievable and appropriate for area source
TEG dehydration units within the UA plus offset and UC boundaries. 
However, we have reduced the reporting requirements for these units as
compared to major sources to lessen the overall burden on these smaller
HAP emission points while still ensuring compliance.

	Comment:  One commenter (IV-D-29) advocated more stringent requirements
for area sources, stating that the proposed requirements were weak and
ineffective and would lead to greater enforcement difficulties and
controversy at the local level.  The commenter was concerned that
industry would fight for area source determinations before local
regulatory agencies.  The commenter noted that local agencies lack the
financial resources and political support to defend against industry. 
The commenter maintained that strengthening the requirements for smaller
HAP sources would force the industry to examine technological control
measures for compliance.  In order for EPA to meet its legal and
substantive requirements under the CAA, the commenter recommended that
EPA should:

require covers, closed vent systems, and control devices with an
emission control efficiency of 95 percent or greater on area source
storage vessels with the potential for flash emissions and an actual
throughput of 500 barrels per day or greater;

require leak detection and repair programs and necessary equipment
modifications for area as well as major sources;

require the same recordkeeping and reporting requirements for major
sources should apply to area sources

MACT or best available control technology (BACT) should be applied to
area sources, instead of GACT; and

implement TEG dehydration unit controls to area sources with a
throughput of 42 thousand m3/day, or greater.  The commenter stated that
EPA offered no real justification for the selected applicability
thresholds.

	Response:  Section 112(d)(5) allows us to set area source standards
according to GACT and management practices.  We believe that we have
properly exercised our discretion and have established appropriate
requirements for sources in this category.

	In addition, there is no basis to regulate storage vessels and
equipment leaks in the final rule, as recommended.  The basis for the
February 1998 proposal was an area source finding which only included
TEG dehydration units.  Subsequently, the July 2005 supplemental
proposal and the final rule are based on the July 1999 listing of oil
and natural gas production facilities under sections 112(c)(3) and
112(k)(3)(B) of the CAA.  This listing was based on information showing
that benzene emissions from TEG dehydration units at area sources of oil
and natural gas production facilities contribute significantly to
nationwide benzene emissions. TEG dehydration units emit a large
majority of the benzene emissions from oil and natural gas production
facilities.  Therefore, the area source category listing focused on
regulating benzene emissions from TEG dehydration units and we did not
include other types of dehydration units or other emission points at
area source oil and natural gas production facilities.  Because we have
not made a risk finding under section112(c)(3) regarding the other
emission points (i.e., storage vessels and equipment leaks), nor
included them in the area source category listing under sections
112(c)(3) and 112(k)(3)(B), we do not have the authority to regulate
them in this final rule.

Control Efficiency

	Comment:  One commenter (IV-D-32) stated that control efficiency
requirements for major and area sources do not both have to be 95
percent.  The commenter remarked that there is no statutory requirement,
nor practical feasibility for such a mandate.

	Response:  Information available to us suggests that there are area
sources that are controlled using condensers.  In addition, we modeled
area source units and found that they could achieve 95 percent control. 
Therefore, we determined that the level of control, when required,
should be 95 percent.

Monitoring, Recordkeeping, and Reporting

	Comment:  Two commenters (IV-D-08 and IV-D-22) stated that the area
source requirements in § 63.775(c) are burdensome and impractical
because some remote locations do not have full time operators onsite. 
Commenter IV-D-08 recommended the following:

develop a generic, simplified contingency plan for area source glycol
dehydrators subject to subpart HH to replace the startup, shutdown, and
malfunction plan codified at §63.10(d)(5);

enable area sources to adopt the contingency plan, referenced above, or
propose their own modifications to it, as part of their notification of
compliance status, as discussed above; and 

allow for compilation of all events in which special action was taken
that is inconsistent with the plan to be submitted in monthly letter
reports.

The commenter expressed appreciation for EPA’s effort to reduce burden
for area sources by not requiring that they have a complete startup,
shutdown, and malfunction plan.

	Response:  As we stated in the July 2005 preamble to the supplemental
proposal (70 FR 39447), in the February 1998 proposal, we proposed only
requiring owners and operators to submit reports of any malfunctions
that are not corrected within 2 calendar days of the malfunction within
7 days of the subject malfunction(s).  It was our intention that owners
or operators would only be required to submit the malfunction reports
and not develop a startup, shutdown, and malfunction (SSM) plan. 
However, as we stated in the July 2005 preamble, we felt that the unique
nature of oil and natural gas production facilities was best addressed
by having owners and operators prepare a SSM plan that would provide the
necessary flexibility in dealing with SSM events at these sites. 
Therefore, the final rule requires SSM plans for sources located within
a UA plus offset and UC boundary.  For sources located outside of these
boundaries, the final rule requires management practices and does not
require startup, shutdown, and malfunction reports for these sources.

REFERENCES

  The commenter has marked up the suggested language for Option 1.

 If EPA extends the area source requirements to rural areas, API would
insert "or in a rural area," here.

 If EPA extends the area source requirements to rural areas, API would
insert "or in a rural area," here.	

 PAGE   

DRAFT

	

  PAGE  iv 

	

DRAFT 

  PAGE  57 

DRAFT   DATE \@ "MMMM d, yyyy"  December 21, 2006 

Table 1.  List of Commenters on the Supplemental Notice of Proposed
Rulemaking on National Emission Standards for Hazardous Air Pollutants: 
Oil and Natural Gas Production Facilities  (continued)

	  PAGE  39 

DRAFT 

DRAFT 

Table 2.  List of Commenters on Proposed Standards for Oil and Natural
Gas Production Industry (continued)a

DRAFT 

.	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.  May 1999.  EPA Document No.
453/R-99-004b.  Section 2.1.3.

.	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.  May 1999.  EPA Document No.
453/R-99-004b.  Section 2.7.

.	National Air Toxics Program:  The Integrated Urban Strategy (64 FR
38705, 38724).

.	Memorandum from Brown, H.P., EC/R Incorporated, to Nizich, G.,
EPA/OAQPS/ESD/WCPG.  June 25, 2005.  Revised Impacts for Area Sources in
the Oil and Natural Gas Production Source Category.  Air Docket No.
OAR-2004-0238-0002 and 0003.

.	Letter from Seitz, J. EPA/OAQPS and E. Schaeffer, EPA/ORE to Director,
Office of Ecosystem Protection, Region I, et al.  December 20, 1999.  p.
3.  Third Extension of January 25, 1995 Potential to Emit Transition
Policy. 

.	Memorandum from Nizich, G., U.S. Environmental Protection Agency, to
Air and Radiation Docket No. EPA-HQ-OAR-2004-0238.  October 18, 2005. 
Discussion with Representatives of the Railroad Commission of Texas
regarding Information Referenced in Comments on the Oil and Natural Gas
Production National Emission Standards for Hazardous Air Pollutants
(NESHAP) Proposal.

.	JFD Consultancy.  Triethylene Glycol Dehydrator Operating Parameters
for Estimating BTEX Emissions.  Prepared for the American Petroleum
Institute.  Air Docket Number A-94-04, Item Number II-D-55.  February
1996.

.	Memorandum from Brown, H.P., EC/R Incorporated, to Nizich, G.,
EPA/OAQPS/SPPD/CCG.  Date to be written. Estimate of the Number of New
TEG Dehydration Units. 

