April 2018

Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Revisions
to C02 Emissions Estimation Methodologies

This memo describes revisions implemented for multiple segments of natural gas and petroleum systems in the
2018 Inventory of U.S. Greenhouse Gas Emissions and Sinks (GHGI). The revisions focus on C02 emissions
calculation methodologies, but for certain sources, both the CH4 and C02 calculation methodologies were revised.
Previous versions of this memo were released in June and October 2017.1,2

The EPA made C02 methodological revisions for sources and segments that already rely on a subpart W-based CH4
emission calculation methodology or where the CH4 calculation methodology was otherwise recently revised. The
subpart W methodology revisions for CH4 emissions estimates are documented in the following memos: the 2014
HF Completion and Workover memo,3 2015 HF Completion and Workover memo,4 2016 Transmission memo,5
2016 Production memo,6 2017 Production memo,7 and 2017 Processing memo.8 The revisions discussed in this
memo create consistency between CH4 and C02 calculation methodologies. In addition, the EPA updated the GHGI
to include both the C02 emissions and the relatively minor CH4 emissions from flare stacks reported under subpart
W in the production and transmission and storage segments.

The sources discussed in this memo include: production segment storage tanks, associated gas venting and
flaring, hydraulically fractured (HF) gas well completions and workovers, production segment pneumatic
controllers, production segment pneumatic pumps, liquids unloading, production segment miscellaneous flaring,
most sources in the gas processing segment, transmission station flares, underground natural gas storage flares,
and transmission and storage pneumatic controllers. The EPA did not consider revisions to the distribution
segment C02 emissions calculation methodology, as discussed in Section 1.2.

1. Background and GHGI Methodology for CO2 Emissions

This section discusses the GHGI methodology for calculating C02 emissions. Section 1.1 describes a C02-to-CH4 gas
content ratio methodology, which is the default approach used in all GHGI segments. This methodology was
applied for numerous sources for the 2017 GHGI, and is still used in the 2018 GHGI for certain sources (excluding
those sources with revisions in section 3). Section 1.2 describes the previous GHGI methodology to calculate C02

1	See https://www.epa.gov/sites/production/files/2017-

06/documents/updates_under_consideration_for_2018_ghgi_emissions_for_co2_from_natu ral_gas_and_petroleum_systems.pdf.

2	See https://www.epa.gOv/sites/production/files/2017-10/documents/2018_ghgi_co2_revisions_under_consideration_2017-10-
25_to_post.pdf.

3	"Overview of Update to Methodology for Hydraulically Fractured Gas Well Completions and Workovers in the Inventory of U.S.
Greenhouse Gas Emissions and Sinks: 1990-2012 (2014 Inventory)," available at https://www.epa.gov/ghgemissions/natural-gas-and-
petroleum-systems-ghg-inventory-updates-1990-2012-inventory-published.

4	"Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2013: Revision to Hydraulically Fractured Gas Well Completions and
Workovers Estimate," available at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems-ghg-inventory-updates-1990-
2013-inventory-published.

5	"Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2014: Revisions to Natural Gas Transmission and Storage Emissions,"
available at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems-ghg-inventory-additional-information-1990-2014-ghg.

6	"Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2014: Revisions to Natural Gas and Petroleum Production Emissions,"
available at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems-ghg-inventory-additional-information-1990-2014-ghg.

7	"Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2015: Revisions to Natural Gas and Petroleum Systems Production
Emissions," available at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems-ghg-inventory-additional-information-
1990-2015-ghg.

8	"Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2015: Revisions to Natural Gas Systems Processing Segment Emissions,"
available at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems-ghg-inventory-additional-information-1990-2015-ghg.

Page 1 of 20

-------
April 2018

emissions for certain sources that relied on emission source-specific methods. The previous GHGI C02 EFs are
documented in Appendix A.

COz-to-CH# Gas Content Ratio Methodology

The default GHGI methodology to calculate C02 emission factors (EFs) relies on CH4 emission factors and an
assumed ratio of C02-to-CH4 gas content. The C02 EF calculation is shown in equation 1:

__	^ /CO, content \

CO-, EF = CH4 EF * I	I	Equation 1

z	4	VCH4 content/	^

The default CH4 and C02 content values for sources in natural gas systems are from the 1996 GRI/EPA study,9
EIA,10 and GTI's Gas Resource Database11 and summarized in Table 1 below.

Table 1. Default Gas Content Values for Natural Gas Systems in the GHGI

Segment

CH4 Content
(vol%)

CO; Content
(vol%)

Production - North East region

78.8

3.04

Production - Mid Central region

0.79

Production - Gulf Coast region

2.17

Production - South West region

3.81

Production - Rocky Mountain region

7.58

Production - West Coast region

0.16

Processing - Before CO2 removal

87.0

3.45

Processing - After CO2 removal

1.0

Transmission and Underground NG Storage

93.4

1.0

LNG Storage and LNG Import/Export

93.4

1.16

Distribution

93.4

1.0

For most of the petroleum production sources evaluated in this memo, the GHGI uses a ratio of C02 to CH4
content, set at 0.017 based on the average flash gas C02 and CH4 content from API TankCalc runs.

The ratio of C02-to-CH4 gas content methodology is used to calculate venting and fugitive C02 EFs, because the
CH4 EFs that are referenced for this methodology represent venting and fugitive emissions, which are
predominantly CH4 with minimal C02 emissions. EPA does not use this methodology in the GHGI to calculate C02
EFs for combustion sources such as flares, for which the inverse is true (C02 is predominant, with minimal CH4
emissions).

1.2 Emission Source-Specific CO2 Calculation Methodologies

The previous GHGI used the following emission source-specific methodologies to calculate C02 emissions from oil
and condensate tanks at production sites, AGR units at natural gas processing plants, and production and
processing flaring.

1.2.1 Oil and Condensate Tanks at Production Sites

The previous GHGI methodology to calculate C02 emissions for oil and condensate tanks used C02 specific EFs.
The EFs were developed using API TankCalc software with varying API gravities. The oil tank EF is the average from

9	Methane Emissions from the Natural Gas Industry, Volume 6: Vented and Combustion Source Summary, Appendix A.

10	U.S. Energy Information Administration. Emissions of Greenhouse Gases in the United States: 1987-1992, Appendix A.

1994.

11	GRI-01/0136 GTI's Gas Resource Database: Unconventional Natural Gas and Gas Composition Databases. Second Edition.
August, 2001.

Page 2 of 20

-------
April 2018

API TankCalc runs for oils with API gravity less than 45, and the condensate tank EF considered data with API
gravity greater than 45. Condensate tank EFs were determined for both controlled and uncontrolled tanks; the
controlled tank EF assumed a control efficiency of 80%. The previous GHGI calculated oil tank C02 emissions by
applying the oil tank emission factor (EF) to 20% of stripper well production and 100% of non-stripper oil well
production. For gas production, the previous GHGI methodology estimated tank emissions by applying the
condensate tank EF to condensate production in each NEMS region.

1.2.2	AGE Units at Natural Gas Processing Plants

The previous GHGI C02 EF for AGR units at natural gas processing plants relied on gas C02 content only. The
difference in the default C02 content before and after C02 removal (3.45% -1.0% = 2.45% of processing plant gas
throughput) is assumed to be emitted.

1.2.3	Flaring

Flaring emissions from the production and processing segments were previously calculated under a single line
item in the production segment of natural gas systems. Therefore, flaring emissions were not specifically
attributed to the natural gas systems processing segment or the petroleum systems production segment. The EF
was based on data from ElA's 1996 greenhouse gas emissions inventory, which estimated the amount of C02
released per BTU of natural gas combusted (0.055 g/BTU). The activity data were annual EIA "Vented and Flared"
gas volumes (MMcf), which are reported under Natural Gas Gross Withdrawals and Production,12 combined with
the estimated national average gas heating value (averaging approximately 1,100 BTU/cf over the time series13).
The EIA Vented and Flared data represents a balancing factor amount that EIA calculates to reconcile reported
upstream and downstream gas volumes, and assumes is potentially emitted to the atmosphere during production
or processing operations; the previous GHGI methodology assumed it was all flared. Details on how much of the
Vented and Flared gas is potentially emitted during natural gas production, petroleum production, and processing
are not available, so the previous GHGI assigned it all to natural gas production. Also, the EIA data do not account
for gas that is flared prior to metering.

Flaring emissions from the transmission and storage segment were not previously calculated in the GHGI. Flaring
emissions from the distribution segment are not currently calculated in the GHGI. Data are unavailable on flaring
emissions in the distribution segment, but they are likely to be insignificant based on the low prevalence of this
activity in the industry segment. EPA did not consider revisions to the distribution segment C02 emissions
calculation methodology for the 2018 GHGI.

2, Available Subpart W Data

Subpart W of the EPA's Greenhouse Gas Reporting Program (GHGRP) collects annual operating and emissions
data on numerous sources from onshore natural gas and petroleum systems that meet a reporting threshold of
25,000 metric tons of C02 equivalent (mt C02e) emissions. Onshore production facilities in subpart W are defined
as a unique combination of operator and basin of operation, a natural gas processing facility in subpart W is each
unique processing plant, a natural gas transmission compression facility in subpart W is each unique transmission
compressor station, an underground natural gas storage facility in subpart W is the collection of subsurface
storage and processes and above ground wellheads, an LNG storage facility in subpart W is the collection of
storage vessels and related equipment, and an LNG import and export facility in subpart W is the collection of
equipment that handles LNG received from or transported via ocean transportation. Facilities in the above-
mentioned industry segments that meet the subpart W reporting threshold have been reporting since 2011;

12	EIA Natural Gas Gross Withdrawals and Production, including the Vented and Flared category, is available at
https://www.eia.gov/dnav/ng/ng_prod_sum_a_EPGO_VGV_mmcf_m.htm

13	EIA Monthly Energy Review. Table A4 - Approximate Heat Content of Natural Gas (Btu per Cubic Feet).

Page 3 of 20

-------
April 2018

currently, six years of subpart W reporting data are publicly available, covering reporting year (RY) 2011 through
RY2016.14

Subpart W activity and emissions data have been used in recent GHGIs to calculate CH4 emissions for several
production, processing, and transmission and storage sources. C02 emissions data from subpart W had not yet
been incorporated into the 2017 GHGI. However, facilities use an identical reporting structure for C02 and CH4.
Therefore, where subpart W CH4 data have been used, the C02 data may be incorporated in a parallel manner.
The 2014 HF Completion and Workover memo, 2016 Transmission memo, 2016 Production memo, 2017
Production memo, and 2017 Processing memo discuss in greater detail the subpart W data available for those
sources.

EPA also reviewed subpart W data that could be used for C02 emission estimates from miscellaneous production
flaring, acid gas removal (AGR) vents, and transmission and storage station flares—sources for which the
emissions were not previously calculated with subpart W data in the GHGI.

Production segment flare emissions are only reported under the "flare stacks" emission source in subpart W if the
flare emissions originate from sources not otherwise covered by subpart W—this emission source is referred to as
"miscellaneous production flaring" for purposes of this memo. Therefore, the subpart W production flares data do
not duplicate flaring emissions reported, for example, under production tank flaring or associated gas flaring. It
also ensures all production flaring emissions are reported for facilities that meet the reporting threshold. Flare
emissions are calculated using a continuous flow measurement device or engineering calculations, the gas
composition, and the flare combustion efficiency. A default flare combustion efficiency of 98% may be applied, if
manufacturer data are not available.

Under subpart W, gas processing facilities calculate AGR unit C02 emissions using one of four methods: (1) C02
CEMS; (2) a vent stream flow meter with C02 composition data; (3) calculation using an equation with the inlet or
outlet natural gas flow rate and measured inlet and outlet C02 composition data; or (4) simulation software (e.g.,
AspenTech HYSYS or API 4679 AMINECalc). CH4 emissions for AGR units are not reported in subpart W.

Transmission and underground natural gas storage stations report emissions from all flaring under the "flare
stacks" emission source as of RY2015. Prior to that, flare emissions reported under subpart W were included in
the reported emissions for the specific source (e.g., reciprocating or centrifugal compressor). Flare emissions are
calculated in subpart W using a continuous flow measurement device or engineering calculations, the gas
composition, and the flare combustion efficiency. A default flare combustion efficiency of 98% may be applied, if
manufacturer data are not available.

3. 2018 GHGI Revisions

For the 2018 GHGI, EPA calculated C02 EFs using the same methodologies that were developed for CH4 EFs in
recent GHGIs. For associated gas venting and flaring and production segment miscellaneous flaring, while there
was an existing methodology, EPA calculated both C02 and CH4 emissions using a revised methodology for the
2018 GHGI. In addition, the EPA updated the GHGI to incorporate subpart W data for C02 from AGR units, and
both the C02 emissions and the relatively minor CH4 emissions from flare stacks in the production and
transmission and storage segments.

14 The GHGRP subpart W data used in the analyses discussed in this memo are those reported to the EPA as of August 5,
2017.

Page 4 of 20

-------
April 2018

duction COz Emlss actors

The EPA developed C02 EFs for several sources in the natural gas and petroleum production segments. The CH4
EFs for oil and condensate tanks, pneumatic controllers, and pneumatic pumps were recently revised using
subpart W data, and EPA applied the same methodology to calculate C02 EFs. There was an existing subpart W-
based CH4 methodology for associated gas venting and flaring and gas well hydraulically fractured completions
and workovers, but a revised methodology was developed for these sources. The EPA also developed a C02
emissions calculation methodology for miscellaneous production flaring. Each of these sources are discussed
below.

3.1.1 Associated Gas Venting and Flaring

The associated gas venting and flaring emissions calculation methodology was revised in the 2017 GHGI to use
subpart W data and calculated CH4 emissions using a national-level, well count-based scaling approach.15
However, stakeholders commented that national-level EFs and AFs would not take into account differences in
associated gas venting and flaring among geographic regions. In particular, over- or under-representation in
GHGRP data by geographic regions where associated gas is vented or flared more or less frequently may
disproportionately contribute to national-level factors. Stakeholders also commented that associated gas
emissions are more directly related to production levels, rather than well counts. In response to stakeholder
comments, the EPA reassessed the associated gas venting and flaring data and finalized a basin-level, production-
based scaling approach for the 2018 GHGI. The final methodology is applied for both C02 and CH4 emissions and is
discussed here. The October 2017 version of this memo documents the national-level approach for C02 (following
the 2017 GHGI methodology) and presents a NEMS region-level, well-based approach that was considered but not
implemented.

The EPA first reviewed the reported subpart W associated gas venting and flaring emissions for RY2011 through
RY2016 to identify basins that contribute the majority of the associated gas emissions. Specifically, if a basin
contributed at least 10 percent of total annual emissions (on a C02 Eq. basis) from associated gas venting and
flaring in any year, then basin-specific EFs and AFs were developed. See Appendix B for the associated gas
emissions by year for all basins. Four basins met this criteria: 220 - Gulf Coast Basin (LA, TX); 360 - Anadarko Basin;
395 - Williston Basin; and 430 - Permian Basin. Associated gas venting and flaring data in all other basins were
combined, and EFs and AFs developed for the other basins as a single group.

EPA calculated EFs for RY2015 and RY2016; subpart W data in earlier years do not contain publicly available
production data. The EPA calculated C02 and CH4 EFs for associated gas venting and flaring by summing the
reported emissions for venting and flaring and dividing by the sum of the reported volume of oil produced during
associated gas venting and flaring. Table 2 and Table 3 present the emissions and oil production data for years
2015 and 2016, and Table 4 shows the resulting EFs. The 2015 EFs were applied to all prior years in the time
series.

Table 2. Associated Gas Venting and Flaring Emissions and Oil Production, Subpart W RY2015

Basin

Venting
CO; (mt)

Venting
CH4 (mt)

Volume of Oil
Produced During
Venting (bbl)

Flaring CO;
(mt)

Flaring CH4
(mt)

Volume of Oil
Produced During
Flaring (bbl)

220 - Gulf Coast Basin (LA, TX)

93

1,259

2,110,981

589,431

2,718

18,591,586

360 - Anadarko Basin

22

906

1,994,628

159,208

695

148,688

395 - Williston Basin

151

1,564

229,586

7,890,206

23,965

264,426,732

430 - Permian Basin

2,675

5,839

5,975,614

2,094,869

8,185

36,912,840

All Other Basins

8,520

6,303

2,522,412

390,300

1,749

27,110,014

15 See the 2017 Production Memo for details.

Page 5 of 20

-------
April 2018

Table 3. Associated Gas Venting and Flaring Emissions and Oil Production, Subpart W RY2016

Basin

Venting
CO; (mt)

Venting
CH4 (mt)

Volume of Oil
Produced During
Venting (bbl)

Flaring CO;
(mt)

Flaring CH4
(mt)

Volume of Oil
Produced During
Flaring (bbl)

220 - Gulf Coast Basin (LA, TX)

267

2,089

1,250,441

298,967

1,187

13,547,580

360 - Anadarko Basin

6

294

175,531

1,185

5

25,735

395 - Williston Basin

140

1,356

234,720

5,035,977

14,017

208,727,344

430 - Permian Basin

216

4,281

4,135,034

1,691,562

6,767

38,294,649

All Other Basins

4,538

4,353

6,711,810

284,496

1,049

18,628,782

Table 4. Calculated Associated Gas Venting and Flaring Emission Factors (kg/bbl/yr)

Basin

Venting CO; EF

Venting CH4 EF

Flaring CO; EF

Flaring CH4 EF

2015

2016

2015

2016

2015

2016

2015

2016

220 - Gulf Coast Basin (LA, TX)

0.04

0.21

0.60

1.67

32

22

0.15

0.09

360 - Anadarko Basin

0.01

0.03

0.45

1.68

1,071

46

4.7

0.20

395 - Williston Basin

0.66

0.60

6.81

5.78

30

24

0.09

0.07

430 - Permian Basin

0.45

0.05

0.98

1.04

57

44

0.22

0.18

All Other Basins

3.38

0.68

2.50

0.65

14

26

0.06

0.08

The EPA calculated two AFs for each basin or group: the percent of oil production with either flaring or venting of
associated gas and, within that subset of production, the fraction that vents and the fraction that flares. The AFs
were calculated for 2015 and 2016, and the 2015 activity factors applied to all prior years. The AF data are
presented in Table 5 and Table 6.

Table 5. Associated Gas Venting and Flaring Production Data and AFs, Subpart W RY2015

Basin

Volume of Oil
Produced
During
Venting (bbl)

Volume of Oil

Produced
During Flaring
(bbl)

Subpart W
Liquids
Production (bbl)

% Production
with Flaring or

Venting of
Associated Gas

%

Production
with
Venting

%

Production
with
Flaring

220 - Gulf Coast Basin (LA, TX)

2,110,981

18,591,586

650,435,832a

3%

10%

90%

360 - Anadarko Basin

1,994,628

148,688

99,146,641

2.2%

93%

7%

395 - Williston Basin

229,586

264,426,732

447,415,171

59%

0.1%

99.9%

430 - Permian Basin

5,975,614

36,912,840

591,656,726

7%

14%

86%

All Other Basins

2,522,412

27,110,014

645,262,423

5%

9%

91%

a. Reported subpart W liquids production exceeded Drillinglnfo production for basin, Drillinglnfo production used to
calculate AF.

Table 6. Associated Gas Venting and Flaring Production Data and AFs, Subpart W RY2016

Basin

Volume of Oil
Produced
During
Venting (bbl)

Volume of Oil

Produced
During Flaring
(bbl)

Subpart W
Liquids
Production (bbl)

% Production
with Flaring or

Venting of
Associated Gas

%

Production
with
Venting

%

Production
with
Flaring

220 - Gulf Coast Basin (LA, TX)

1,250,441

13,547,580

516,246,773a

3%

8%

92%

360 - Anadarko Basin

175,531

25,735

94,789,700

0.2%

87%

13%

395 - Williston Basin

234,720

208,727,344

322,617,029

65%

0.1%

99.9%

430 - Permian Basin

4,135,034

38,294,649

533,358,906

8%

10%

90%

All Other Basins

6,711,810

18,628,782

1,464,067,958a

2%

26%

74%

a. Subpart W liquids production exceeded Drillinglnfo production for basin, Drillinglnfo production used to calculate AF.

EPA uses total liquids production data for each basin or group to calculate national emissions. Total liquids
production data for each basin were determined from Drillinglnfo, while the total national liquids production was

Page 6 of 20

-------
April 2018

available from EIA (consistent with current methodologies for other GHGI sources that rely on total national
production data). Therefore, the national production for all other basins equals the EIA production minus the
Drillinglnfo production for each of the four basins. The total liquids production data for 2015 and 2016 are
provided in Table 7, and the resulting national emissions are shown in Table 8.

Table 7. Total Liquids Production (bbl), by Basin

Basin

Year 2015

Year 2016

220 - Gulf Coast Basin (LA, TX)

650,435,832

516,246,773

360 - Anadarko Basin

144,644,537

122,734,407

395 - Williston Basin

456,423,760

396,753,744

430 - Permian Basin

688,208,748

733,002,118

All Other Basins

1,494,207,123

1,464,067,958

Table 8. Calculated Total Associated Gas Venting and Flaring Emissions

Basin

Venting CO> (mt)

Venting CH4 (mt)

Flaring CO; (mt)

Flaring CH4 (mt)

2015

2016

2015

2016

2015

2016

2015

2016

220 - Gulf Coast
Basin (LA, TX)

93

267

1,259

2,089

589,431

298,967

2,718

1,187

360 - Anadarko Basin

31

8

1,321

381

232,268

1,534

1,014

7

395 - Williston Basin

154

173

1,596

1,668

8,049,073

6,193,234

24,447

17,238

430 - Permian Basin

3,112

297

6,792

5,883

2,436,729

2,324,735

9,520

9,301

All Other Basins

19,728

4,538

14,596

4,353

903,802

284,496

4,049

1,049

Total

23,119

5,282

25,564

14,375

12,211,303

9,102,967

41,749

28,782

3.1.2 Miscellaneous Production Flaring

The EPA used subpart W RY2015 and RY2016 miscellaneous production flaring (reported under "flare stacks")
emissions data to revise the GHGI and more fully account for flare emissions in the production segment. Subpart
W data for this source were not previously considered. The EPA calculated the C02 and CH4 EFs using a national-
level, well count-based scaling approach for the 2018 GHGI public review draft; this methodology is documented
in the previous July and October 2017 versions of this memo. However, similar to associated gas venting and
flaring, stakeholders recommended a basin-level, production-based scaling approach. After evaluating the data, a
basin-level, production-based scaling approach was applied for the 2018 GHGI, and is documented here.

The EPA reviewed the reported subpart W miscellaneous production flaring emissions for RY2011 through RY2016
to identify basins that contribute the majority of the associated gas emissions. Specifically, if a basin contributed
at least 10 percent of total annual emissions (on a C02 Eq. basis) from miscellaneous production flaring in any
year, then basin-specific emission factors and activity factors were developed. See Appendix C for the
miscellaneous production flaring emissions by year for all basins. Three basins met this criteria: 220 - Gulf Coast
Basin (LA, TX); 395 - Williston Basin; and 430 - Permian Basin. Miscellaneous production flaring data in all other
basins were combined, and EFs and AFs developed for the other basins as a single group. EFs and AFs were
developed using RY2015 and RY2016 data, as prior years do not contain publicly available production data.

Miscellaneous production flaring emissions are not reported separately for gas and oil production. Therefore, to
use reported emissions data for separate natural gas and petroleum systems GHGI estimates, the EPA calculated
the fraction of wells that were gas and oil wells for each facility, using the well counts reported in the Equipment
Leaks section of subpart W.16 The EPA then apportioned each facility's reported miscellaneous production flaring

16 Three facilities with miscellaneous production flaring emissions did not report well counts. Therefore, for these three
facilities, the EPA determined the fraction of sub-basins applicable to gas production (i.e., sub-basins with high permeability
gas, shale gas, coal seam, or other tight reservoir rock formation types) and oil production (i.e., sub-basins with the oil
formation type), and applied these fractions in the calculations.

Page 7 of 20

-------
April 2018

C02 and CH4 emissions by production type, and summed the facility-level C02 and CH4 emissions for each
production type to the basin-level to estimate total reported miscellaneous flaring C02 and CH4 emissions from
natural gas and oil production, for each basin or group.

Next, EPA used gas and liquids production data to develop EFs for calculating the national total emissions. The
EPA calculated EFs by dividing the basin-level C02 and CH4 emissions for natural gas and oil production by the
summation of the reported gas produced from wells (for natural gas production EFs) and liquids produced (for oil
production EFs). These emissions data, production data, and calculated EFs are provided in Table 9 through Table
12 below. The 2015 EFs were applied to all prior years in the time series.

Table 9. GHGRP Subpart W RY2015 Natural Gas Production C02 and CH4 Emissions and Activity
Data and Calculated EFs for Miscellaneous Production Flaring

Basin

Gas CO;
(mt)

Gas CH4
(mt)

Gas Produced from
Wells (mscf)

Gas CO; EF
(kg/mscf/yr)

Gas CH4 EF
(kg/mscf/yr)

220 - Gulf Coast Basin (LA, TX)

324,079

1,157

3,161,594,496

1.03E-01

3.66E-04

395 - Williston Basin

56

0

645,705,949a

8.61E-05

3.14E-07

430 - Permian Basin

673,592

2,992

2,367,810,821a

2.84E-01

1.26E-03

All Other Basins

310,453

1,337

20,352,492,312a

1.53E-02

6.57E-05

a. Subpart W production exceeded Drillinglnfo production for basin, Drillinglnfo production used.

Table 10. GHGRP Subpart W RY2015 Oil Production C02 and CH4 Emissions and Activity Data and
Calculated EFs for Miscellaneous Production Flaring

Basin

Oil CO; (mt)

Oil CH4 (mt)

Liquids Produced
(bbl)

Oil CO; EF
(kg/bbl/yr)

Oil CH4 EF
(kg/bbl/yr)

220 - Gulf Coast Basin (LA, TX)

859,858

3,548

652,726,411a

1.32E+00

5.44E-03

395 - Williston Basin

856,957

2,145

447,415,171

1.92E+00

4.79E-03

430 - Permian Basin

424,156

1,626

591,656,726

7.17E-01

2.75E-03

All Other Basins

540,935

1,861

743,813,115a

7.27E-01

2.50E-03

a. Subpart W production exceeded Drillinglnfo production for basin, Drillinglnfo production used.

Table 11. GHGRP Subpart W RY2016 Natural Gas Production C02 and CH4 Emissions and Activity
Data and Calculated EFs for Miscellaneous Production Flaring

Basin

Gas CO;
(mt)

Gas CH4
(mt)

Gas Produced from
Wells (mscf)

Gas CO; EF
(kg/mscf/yr)

Gas CH4 EF
(kg/mscf/yr)

220 - Gulf Coast Basin (LA, TX)

213,698

584

2,661,846,306

8.03E-05

2.19E-07

395 - Williston Basin

206

0

649,228,154a

3.18E-07

5.28E-10

430 - Permian Basin

438,567

1,939

2,356,640,169

1.86E-04

8.23E-07

All Other Basins

339,247

1,573

19,553,610,690a

1.73E-05

8.05E-08

a. Subpart W production exceeded Drillinglnfo production for basin, Drillinglnfo production used.

Table 12. GHGRP Subpart W RY2016 Oil Production C02 and CH4 Emissions and Activity Data and
Calculated EFs for Miscellaneous Production Flaring

Basin

Oil CO; (mt)

Oil CH4 (mt)

Liquids Produced
(bbl)

Oil CO; EF
(kg/bbl/yr)

Oil CH4 EF
(kg/bbl/yr)

220 - Gulf Coast Basin (LA, TX)

389,281

1,630

518,218,649"

7.51E-04

3.15E-06

395 - Williston Basin

274,154

778

322,617,029

8.50E-04

2.41E-06

430 - Permian Basin

563,672

1,991

533,358,906

1.06E-03

3.73E-06

All Other Basins

414,762

1,035

689,536,735a

6.02E-04

1.50E-06

a. Subpart W production exceeded Drillinglnfo production for basin, Drillinglnfo production used.

Page 8 of 20

-------
April 2018

EPA calculated national emissions using the appropriate national production (i.e., total gas production or liquids
production) for each basin or group. Total gas production data for each basin and for the nation were determined
from Drillinglnfo. Total liquids production data for each basin were determined from Drillinglnfo, while the total
national liquids production was available from EIA (consistent with current methodologies for other GHGI sources
that rely on total national production data). Therefore, the national liquids production for all other basins equals
the EIA production, minus the Drillinglnfo production for each of the three basins. The production data and
resulting national emissions for 2015 and 2016 are shown in Table 13 and Table 14.

Table 13. Total Production Data and Miscellaneous Production Flaring Emissions for Natural Gas

and Petroleum Systems, Reporting Year 2015

Basin

Total Gas
Production
(mscf)

Total Liquids
Production
(bbl)

Gas CO;
(mt)

Gas CH4
(mt)

Oil CO;
(mt)

Oil CH4
(mt)

220 - Gulf Coast Basin (LA, TX)

3,519,664,923

652,726,411

360,782

1,288

859,858

3,548

395 - Williston Basin

645,705,949

456,442,746

56

0

874,248

2,188

430 - Permian Basin

2,367,810,821

688,752,179

673,592

2,992

493,763

1,893

All Other Basins

24,940,124,177

1,635,998,664

380,431

1,639

1,189,773

4,094

Total

31,473,305,870

3,433,920,000

1,414,861

5,918

3,417,643

11,724

Table 14. Total Production Data and Miscellaneous Production Flaring Emissions for Natural Gas

and Petroleum Systems, Reporting Year 2016

Basin

Total Gas
Production
(mscf)

Total Liquids
Production
(bbl)

Gas C02
(mt)

Gas CH4
(mt)

Oil C02
(mt)

Oil CH4
(mt)

220 - Gulf Coast Basin (LA, TX)

3,061,920,423

518,218,649

245,817

672

389,281

1,630

395 - Williston Basin

649,228,154

396,772,982

206

0

337,170

957

430 - Permian Basin

2,546,961,000

733,544,659

473,985

2,095

775,235

2,738

All Other Basins

23,551,484,913

1,584,268,710

408,609

1,895

952,951

2,378

Total

29,809,594,491

3,232,805,000

1,128,617

4,662

2,454,637

7,703

3.1.2 Production Tanks

Based on the CH4 EF methodology documented in the 2017 Production memo, the EPA calculated oil and
condensate tank C02 EFs for several tank categories, using subpart W data: large tanks with flaring; large tanks
with a vapor recovery unit (VRU); large tanks without controls; small tanks with flaring; small tanks without
flaring; and malfunctioning separator dump valves. EPA applied several steps described in the 2017 Production
memo to apportion the reported subpart W data to each of the categories. EPA then summed the emissions and
divided by the throughput for each tank category. Table 15 presents the resulting C02 EFs for RY2015 (which are
applied for 2015 and all prior years in the time series) and RY2016.

Table 15. GHGRP Subpart W-based Oil and Condensate Tank C02 EFs (kg/bbl/yr)

Tank Category

Oil Tanks EF

Condensate Tanks EF

2015

2016

2015

2016

Large Tanks with Flaring

7.21

6.98

8.33

10.90

Large Tanks with VRU

0.037

0.025

0.11

0.12

Large Tanks without Controls

0.016

0.019

0.019

0.026

Small Tanks with Flaring

0.27

1.64

1.96

4.46

Small Tanks without Flares

0.078

0.066

0.28

0.46

Malfunctioning Dump Valves

0.013

0.012

8.19E-05

6.98E-05

Page 9 of 20

-------
April 2018

3.1.4 Pneumatic Controllers

Based on the CH4 EF methodology documented in the 2016 Production memo, the EPA calculated pneumatic
controller C02 EFs for low, intermittent, and high bleed controllers using Subpart W RY2014 data. EPA divided the
reported emissions by the number of reported controllers for each controller type to calculate EFs. All pneumatic
controllers data were considered together, and thus pneumatic controller EFs for natural gas and petroleum
systems are identical. Table 16 presents the subpart W activity and emissions data, along with the calculated C02
EFs.

Table 16. GHGRP Subpart W RY2014 Activity and Emissions Data and Calculated EFs for Pneumatic Controllers

Controller Type

# Controllers

Total CO;
Emissions (mt)

CO; EF
(kg/controller/yr)

Low Bleed

198,941

2,382

12

Intermittent Bleed

561,283

98,269

175

High Bleed

27,208

9,790

360

3.1.5 Pneumatic Pumps

Based on the CH4 EF methodology documented in the 2016 Production memo, the EPA calculated a pneumatic
pump C02 EF using Subpart W RY2014 data. EPA divided the reported emissions by the number of reported
pneumatic pumps to calculate the EF. All pneumatic pumps data were considered together, and thus the EF for
natural gas and petroleum systems is identical. Table 17 presents the subpart W activity and emissions data, along
with the calculated C02 EF.

Table 17. GHGRP Subpart W RY2014 Activity and Emissions Data and Calculated EF for Pneumatic Pumps

# Pumps

Total CO;
Emissions (mt)

CO; EF
(kg/pump/yr)

79,760

11,647

146

3.1.3 IIF Gas Well Completions and Workovers

EPA calculated C02 emissions for the 2018 GHGI public review draft using the CH4 EF methodology documented in
the 2014 HF Completion and Workover memo and 2015 HF Completion and Workover memo. See the earlier
versions of this memo from June and October 2017 for the resulting EFs. However, both the C02 and CH4
calculation methodologies for this source were revised for the final 2018 GHGI, and these revisions are
documented in the 2018 Year-Specific Emissions memo.17

3.1.6 Liquids Unloading

EPA calculated C02 emissions for the 2018 GHGI public review draft using the CH4 EF methodology documented in
the 2017 Production memo. See the earlier versions of this memo from June and October 2017 for the resulting
EFs. However, both the C02 and CH4 calculation methodologies for this source were revised for the final 2018
GHGI, and these revisions are documented in the 2018 Year-Specific Emissions memo.

3.2 cessing CO2 Emission Factors

The EPA developed gas processing C02 EFs for the plant grouped emission sources (reciprocating compressors,
centrifugal compressors with wet seals, centrifugal compressors with dry seals, dehydrators, flares, and plant
fugitives), blowdowns and venting, and AGR vents. The CH4 EFs for the grouped sources and blowdowns and
venting were recently revised using subpart W data, and the EPA applied the same methodology to calculate C02
EFs. AGR vent emissions were not previously calculated from subpart W data (as CH4 emissions are not reported
for this source), but the EPA calculated a subpart W-based C02 EF and determined the corresponding activity data
for this source.

17 "Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Revisions to Create Year-Specific Emissions and Activity Factors,"
available online at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems-ghg-inventory-additional-information-1990-
2016-ghg.

Page 10 of 20

-------
April 2018

Based on the CH4 EF methodology documented in the 2017 Processing memo, the EPA calculated the plant
grouped source C02 EFs using subpart W data (the purpose of the plant grouped EF is discussed in Section 3.4).
Subpart W RY2015 and RY2016 data and calculated C02 EFs for the plant grouped sources are presented in Table
18 and Table 19.

Table 18. GHGRP Subpart W RY2015 Emissions and Activity Data and Calculated EFs for Gas

Processing Plant Grouped Sources

Emission Source

CO;
Emissions
(mt)

Activity Count (plants
or compressors)

CO; EF
(kg/compressor/yr
or kg/plant/yr)

Reciprocating compressors

7,618

2,678

compressors

2,845

Centrifugal compressors with wet seals

1,259

264

compressors

4,768

Centrifugal compressors with dry seals

21

215

compressors

400

Dehydrators

7,430

466

plants

15,944

Flares

4,231,009

466

plants

9,079,418

Plant fugitives

2,244

466

plants

4,816

Plant Grouped Sources

4,249,580

466

plants

9,119,411

Table 19. GHGRP Subpart W RY2016 Emissions and Activity Data and Calculated EFs for Gas

Processing Plant Grouped Sources

Emission Source

CO;
Emissions
(mt)

Activity Count (plants
or compressors)

CO; EF
(kg/compressor/yr
or kg/plant/yr)

Reciprocating compressors

7,275

2,737

compressors

2,658

Centrifugal compressors with wet seals

839

226

compressors

3,711

Centrifugal compressors with dry seals

39

228

compressors

474

Dehydrators

4,467

447

plants

9,994

Flares

3,621,791

447

plants

8,102,440

Plant fugitives

2,599

447

plants

5,813

Plant Grouped Sources

3,637,009

447

plants

8,136,640

Based on the CH4 EF methodology documented in the 2017 Processing memo, the EPA also calculated the
blowdown and venting C02 EF using subpart W data. Subpart W RY2015 data and the calculated C02 EF for
blowdowns and venting are presented in Table 20.

Table 20. GHGRP Subpart W RY2015 Emissions and Activity Data and Calculated EF for Gas

Processing Blowdown and Venting

RY

C02 Emissions
(mt)

Activity Count
(plants)

CO; EF
(kg/plant/yr)

2015

11,059

466

23,731

2016

7,817

447

17,487

For AGR vent emissions, the existing CH4 EF methodology does not rely on subpart W, but the EPA applied a
similar methodology as the other processing sources to develop C02 EFs and activity data from subpart W data.
The EPA summed the reported AGR vent C02 emissions for gas processing plants and divided by the total reported
count of plants for each RY to calculate C02 EFs. Subpart W RY2015 and RY2016 data and the calculated C02 EFs
for AGR vents are presented in Table 21. To calculate national C02 emissions, the C02 EF was multiplied by the
number of gas plants each year.

Page 11 of 20

-------
April 2018

Table 21. GHGRP Subpart W Emissions and Activity Data and Calculated EFs for Gas Processing AGR Vents

Year

CO; Emissions
(mt)

Activity Count
(plants)

CO; EF
(kg/plant/yr)

2015

10,441,754

466

22,407,197

2016

11,101,161

447

24,834,813

3.3 Transmission and Storaj	nission Factors

3,3,1 Pneumatic Controllers

Based on the CH4 EF methodology documented in the 2016 Transmission memo, the EPA calculated transmission
station and storage station pneumatic controller C02 EFs for low, intermittent, and high bleed controllers using
Subpart W RY2011 - RY2016 data. The EPA divided the reported emissions by the number of reported controllers
for each controller type to calculate EFs. Table 22 and Table 23 present the subpart W activity and emissions data,
along with the calculated C02 EFs. The RY2011 EFs were applied for all prior years in the time series.

Table 22. GHGRP Subpart W Activity and Emissions Data and Calculated EFs for Transmission Station

Pneumatic Controllers

Controller Type

Data Element

2011

2012

2013

2014

2015

2016



Total Count

2,203

1,114

1,158

1,173

1,508

1,000

High Bleed

C02 Emissions (mt)

203

106

106

107

121

85



CO2 EF (kg/controller/yr)

92

95

91

91

80

85



Total Count

8,343

9,114

9,903

11,160

10,891

11,122

Intermittent Bleed

CO2 Emissions (mt)

673

736

747

134

105

120



CO2 EF (kg/controller/yr)

81

81

75

12

10

11



Total Count

644

880

857

1,078

1,033

943

Low Bleed

CO2 Emissions (mt)

4.6

6.2

6.2

6.7

4.3

4.5



CO2 EF (kg/controller/yr)

7.1

7.0

7.3

6.2

4.2

4.8

Table 23. GHGRP Subpart W Activity and Emissions Data and Calculated EFs for Underground Natural



Gas Storage Station Pneumatic Controllers







Controller Type

Data Element

2011

2012

2013

2014

2015

2016



Total Count

1,253

1,100

1,089

1,271

1,024

1,051

High Bleed

CO2 Emissions (mt)

116

118

116

117

64

97



CO2 EF (kg/controller/yr)

92

107

106

92

63

92



Total Count

1,391

1,539

1,601

2,045

2,098

2,288

Intermittent Bleed

CO2 Emissions (mt)

16

21

21

24

22

50



CO2 EF (kg/controller/yr)

12

13

13

12

10

22



Total Count

250

319

366

319

320

289

Low Bleed

CO2 Emissions (mt)

1.9

2.4

2.8

2.2

1.4

1.6



CO2 EF (kg/controller/yr)

7.5

7.4

7.6

7.0

4.4

5.5

3,3,2 Flares

The EPA developed GHGI flare C02 EFs for transmission stations and underground natural gas storage using
subpart W data. As discussed in Section 1.3, the GHGI C02 emissions calculation methodology did not previously
calculate C02 emissions from flares. Therefore, the EPA updated the methodology to calculate C02 emissions with
new line items for transmission and storage flares.

Page 12 of 20

-------
April 2018

The EPA divided the reported flare C02 and CH4 emissions by the number of reported stations to calculate the EFs.
Subpart W transmission station and underground natural gas storage flare data are presented in Table 24 and
Table 25. The applicable activity data to calculate national emissions are the national number of stations, which
are already calculated in the GHGI. The RY2015 EFs were applied for all prior years in the time series.

Note these flaring emissions estimates were developed from reported GHGRP data, wherein transmission
compressor stations that service underground storage fields might be classified as transmission compression as
the primary function. Therefore, a fraction of the transmission station flaring emissions may occur at stations that
service storage facilities; such stations typically require flares, compared to a typical transmission compressor
station used solely for mainline compression that does not require liquids separation, dehydration, and flaring.

Table 24. GHGRP Subpart W RY2015 Emissions and Activity Data and Calculated EFs for

Transmission Station Flares

Year

Total #
Stations

# Stations
With Flares

# Flares

Total CO;
Emissions (mt)

CO; EF
(kg/station/yr)

Total CH4
Emissions (mt)

CH4 EF
(kg/station/yr)

2015

524

18

30

23,833

45,483

93

177

2016

529

17

26

25,116

47,479

112

212

Table 25. GHGRP Subpart W RY2015 Emissions and Activity Data and Calculated EFs for
Underground Natural Gas Storage Flares

Year

Total #
Stations

# Stations
With Flares

# Flares

Total CO;
Emissions (mt)

CO; EF
(kg/station/yr)

Total CH4
Emissions (mt)

CH4 EF
(kg/station/yr)

2015

53

9

23

3,587

67,676

35

651

2016

53

9

21

2,343

44,214

30

572

Time Series Considerations

For the production segment sources discussed in Section 3.1, in general, the EPA applied the same methodology
to calculate C02 over the time series as used for calculating CH4 emissions over the time series.18 For oil and
condensate tanks, the EPA applies category-specific EFs for every year of the time series and for pneumatic
controllers and pumps, category-specific EFs are applied for each year of the time series.

For associated gas venting and flaring, for CH4, the EPA applied the subpart W 2015 EFs for years prior to 2015 and
year-specific subpart W EFs were applied for 2015 and forward.

For the miscellaneous production flaring time series, the previous GHGI flare emission estimate (representing
both production and processing), fluctuated based on activity data (ElA's estimated annual vented and flared
volumes). Assessment of subpart W C02 data over the time series for this source indicates that miscellaneous
production flaring emissions do not show a clear trend. See the Requests for Stakeholder Feedback section for
more information. In the revised approach to use subpart W-based EFs (kg/mscf or kg/bbl), the EF was held
constant for years prior to 2015 and flare emission estimates fluctuated with gas and liquids production data over
the time series.

For certain processing sources discussed in Section 3.2, the EPA applied the same methodology to calculate C02
over the time series as used for calculating CH4 emissions over the time series.19 For plant grouped emission
sources and blowdowns and venting, GRI/EPA 1996 EFs are used for 1990 through 1992; EFs calculated from
subpart W are used for 2011 forward; and EFs for 1993 through 2010 are developed through linear interpolation.

18	Additional details on current time series calculations for production segment sources are provided in the 2014 HF
Completion and Workover memo, 2015 HF Completion and Workover memo, 2016 Production memo, and 2017 Production
memo.

19	Additional details on current time series calculations are provided in the 2017 Processing memo.

Page 13 of 20

-------
April 2018

For C02 from AGR vents, the EPA adopted a similar methodology as the other processing sources (maintain the
current GRI/EPA 1996 EFs for 1990 through 1992, apply the subpart W-based EFs for 2011 forward, and develop
EFs for 1993 through 2010 using linear interpolation).

For transmission and storage flares, the EPA applied the 2015 subpart W-based EF (kg/station) for all prior years
of the time series and year-specific EFs for 2015 and forward.

4. National Emissions Estimates

For sources with the largest contribution to C02 emissions (e.g., flaring sources), national C02 emissions for year
2015 using the subpart W-based approaches discussed in Section 3 (and implemented in the 2018 GHGI) are
compared against the 2017 GHGI in Table 26 and Table 27.

Table 26. Natural Gas Systems Year 2015 National C02 Emissions (MMT) for 2018 GHGI

Compared to 2017 GHGI



2017 GHGI

201S GHGI







Exploration

NA

0.29

HF Completions

0.07

0.28

Production

17.6

3.40

Miscellaneous Flaring

17.6a

1.41

Tanks

0.03

1.09

HF Workovers

0.03

0.08

Processing

23.7

21.04

AGR Vents

23.6

14.95

Plant Grouped Sources

0.1

6.08

Transmission & Storage

0

0.15

Transmission Flares

0

0.11

Underground Storage Flares

0

0.02

Distribution

0

0.01





1	

NA (Not Applicable)

a. Also represents flaring from petroleum production and gas processing.

Table 27. Petroleum Systems Year 2015 National C02 Emissions (MMT) for 2018 GHGI Compared

to 2017 GHGI







Exploration

NA

0.26

Well Testing

NE

0.26

Production

0.64

24.48

Associated Gas

NE

12.23

Tanks

0.52

8.72

Miscellaneous Flaring

NEa

3.42

Transportation

NE

NE

Refining

2.93

4.01













NA (Not Applicable)

NE (Not Estimated)

a. In the 2017 GHGI, emissions were generally included within the natural gas

systems production flaring estimate.

Page 14 of 20

-------
April 2018

The C02 revisions resulted in an overall shift of C02 emissions from Natural Gas systems to Petroleum systems.

This is due to the availability of industry segment-specific and emission source-specific data in subpart W, whereas
previous data sources were not as granular. The previous GHGI accounted for all onshore production and gas
processing flaring emissions under a single line item in the production segment of natural gas systems. Using the
revised approaches, flaring emissions are now specifically calculated for natural gas production, petroleum
production, and gas processing (within the plant grouped emission sources). The shift in C02 emissions from
Natural Gas systems to Petroleum systems is also due to the inclusion of associated gas flaring as a specific line
item under Petroleum systems; this is the largest source of C02 emissions for the revisions.

5. October 2017 Requests for Stakeholder Feedback

The EPA initially sought feedback on the questions below in the version of this memo released in October 2017.
The questions below were minimally altered to specifically cite the October 2017 memo. The EPA discusses
feedback received, and further planned improvements to the GHGI methodology, in Chapter 3.8 of the Inventory
of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 (April 2018). The EPA continues to welcome additional
stakeholder feedback on these questions for potential updates to future GHGIs.

General

1.	EPA seeks stakeholder feedback on the general approach of using subpart W reported C02 emissions data
to revise the current C02 emissions calculation methodology (described in Section 1) in the GHGI.

2.	EPA seeks feedback on using consistent calculation methodologies for both CH4 and C02, when GHGI
relies on subpart W data. Are there sources where the CH4 and C02 methodologies based on subpart W
should differ?

Associated Gas Venting and Flaring (Section 3.1.1)

3.	EPA seeks feedback on the methodology to calculate national emissions from associated gas venting and
flaring. In particular, which methodology discussed in Section 3.1.1 of the October 2017 memo (national-
level, or NEMS region-level) or other approach best reflects national-level emissions from associated gas
venting and flaring by taking into account variability of this source?

4.	What scale-up assumptions should EPA make regarding associated gas venting or flaring for regions that
do not report any oil well data to GHGRP? Should EPA assume that these regions have no such activity, or
should EPA assign surrogate EF and AF values (e.g., average from all other reported regions, or some
other methodology)?

5.	Should EPA consider an approach not presented in Section 3.1.1 of the October 2017 memo?

a.	For example, scaling subpart W-based estimates using production rather than oil well counts?

b.	For example, disaggregating to the AAPG basin-level?

GHGI Sources that Are Not Currently Estimated Using Subpart W data

6.	In the October 2017 memo, Section 3.1.7 discusses considerations for developing EFs and associated
activity data for miscellaneous production flaring that facilitate scaling reported subpart W data to a
national level. The EPA has presented a preliminary approach that develops an EF in units of emissions per
well. National active well counts would be paired with such EF to calculate emissions in the GHGI. The EPA
seeks feedback on this approach, or suggestions of other approaches that would facilitate scaling to a
national level and time series population.

Page 15 of 20

-------
April 2018

7. For sources discussed in the October 2017 memo that do not currently estimate CH4 emissions using
subpart W, EPA is considering which year(s) of subpart W data to use in developing the C02 emissions
methodologies. For miscellaneous production flaring, the EPA reviewed reported emissions and activity
data for RY2011 - RY2014. However, wellhead counts for RY2011 - RY2014 are only reported by those
facilities that calculated equipment leak emissions using Methodology 1, and as such, are not
comprehensive. At the time of the 2016 Production memo, 83% of reporting facilities for RY2011, 85% of
RY2012 reporting facilities, 93% of RY2013 facilities, and 98% of RY2014 reporting facilities reported
wellhead counts under Methodology 1. In addition, facilities only reported total wellheads and did not
report gas and oil wellhead counts separately for RY2011 - RY2014. The EPA calculated the C02 EFs under
consideration using RY2015 only, because well counts from all reporting facilities are reported. However,
the EPA requests feedback on whether it is appropriate to consider data from prior reporting years, which
would have more uncertainty due to incomplete coverage, in order to show a trend over the time series.
Table 28 provides the reported subpart W emissions and activity data for RY2011-RY2015.

Table 28. GHGRP Subpart W Emissions and Activity Data for Miscellaneous Production Flaring

Year

CO: Emissions
(mt)

# Flares

# Wells (a)

CO; EF
(kg/well)

2011

2,252,297

13,509

371,604

6,061

2012

3,616,326

16,356

398,137

9,083

2013

4,596,329

21,098

415,355

11,066

2014

4,841,116

22,155

502,391

9,636

2015

3,779,110

20,293

527,170

7,169

a. Total gas and oil wellheads. Wellhead counts for RY2011 through RY2014 are
available from those onshore production facilities that calculated equipment leak
emissions using Methodology 1.

For transmission and storage segment flares, the EPA relies on RY2015 data for the revisions under
consideration, because all flaring emissions are reported under the flare stacks source. Whereas, for
RY2011 - RY2014, flare emissions are reported under flare stacks and each individual emission source.

8. Section 3.4 in the October 2017 memo discusses time series considerations for transmission and storage
flares. The EPA is considering applying a subpart W-based EF (kg/station) for all years of the time series.
However, few transmission and storage stations reported flares for RY2015 (see Table 24 and Table 25).
Therefore, EPA might alternatively assume that flares did not operate in 1990 (i.e., an EF of 0), apply the
subpart W-based EF for 2011 forward, and apply linear interpolation from 1991 through 2010. The EPA
seeks feedback on these approaches, or suggestions of other approaches to time series population.

Page 16 of 20

-------
April 2018

Appendix	rent (2017) GHGI COz Emission Factors

All EFs are presented in the same units as the EFs under consideration for the 2018 GHGI; kg/[unit].

	1	1	

Natural Gas & Petroleum Production

Stripper Wells (for Associated Gas Venting)

2.47

kg/well

Condensate Tank Vents - Without Control Devices

0.18

kg/bbl

Condensate Tank Vents - With Control Devices

0.037

kg/bbl

Oil Tanks

0.18

kg/bbl

HF Gas Well Completions and Workovers

18,367a

kg/event

Pneumatic Controllers, all bleed types (Natural Gas)

144a

kg/controller

Low Bleed Pneumatic Controllers (Petroleum)

8.8

kg/controller

Intermittent Bleed Pneumatic Controllers (Petroleum)

83.9

kg/controller

High Bleed Pneumatic Controllers (Petroleum)

238.9

kg/controller

Pneumatic Pumps (Natural Gas)

168.4a

kg/pump

Pneumatic Pumps (Petroleum)

82.8

kg/pump

Liquids Unloading with Plunger Lifts

613a

kg/well

Liquids Unloading without Plunger Lifts

678a

kg/well

Onshore Production & Processing - Flaring Emissions

40,624

kg/well

Natural Gas Processing

Reciprocating compressors - before C02 removal

4,764

kg/compressor

Reciprocating compressors - after C02 removal

1,058

kg/compressor

Centrifugal compressors with wet seals - before C02 removal

21,859

kg/compressor

Centrifugal compressors with wet seals - after C02 removal

4,854

kg/compressor

Centrifugal compressors with dry seals - before C02 removal

10,719

kg/compressor

Centrifugal compressors with dry seals - after C02 removal

2,380

kg/compressor

Plant fugitives - before C02 removal

3,364

kg/plant

Plant fugitives - after C02 removal

747

kg/plant

Kimray pumps

859

kg/plant

Dehydrator vents

5,291

kg/plant

Plant Grouped Sources

95,303

kg/plant

AGR vents

35,394,396

kg/plant

Blowdowns and venting

8,363

kg/plant

Transmission

High Bleed Pneumatic Controllers

84.43

kg/controller

Intermittent Bleed Pneumatic Controllers

10.95

kg/controller

Low Bleed Pneumatic Controllers

6.22

kg/controller

Underground NG Storage

High Bleed Pneumatic Controllers

82.21

kg/controller

Intermittent Bleed Pneumatic Controllers

10.74

kg/controller

Low Bleed Pneumatic Controllers

6.34

kg/controller

a. Average EF based on data from all NEMS regions.

Page 17 of 20

-------
April 2018

Appendix B - GHGRP Subpart W Associated Gas Venting and Flaring Emissions, by basin, for RY2011-2016



2011

2012

2013

2014

2015

2016

Basin

Total CO' +
CH, Emissions

% of
Total

Total CO' +
CH, Emissions

% of
Total

Total CO' +
CH, Emissions

% of
Total

Total CO' +
CH, Emissions

% of
Total

Total CO' +
CH, Emissions

% of
Total

Total CO' +
CH, Emissions

% of
Total



(mt CO e)

(mt CO'e)

(mt CO'e)

(mt CO'e)

(mt CO'e)

(mt CO'e)

160 - Appalachian Basin

256

0%

267

0%

272

0%

208

0%

183

0%

6,061

0%

160A - Appalachian Basin (Eastern
Overthrust Area)

16,224

0%

23,477

0%

27,119

0%

15,055

0%

36,404

0%

18,016

0%

210 - Mid-Gulf Coast Basin

22,825

0%

14,535

0%

32,584

0%

68,569

1%

95,521

1%

103,081

1%

220 - Gulf Coast Basin (LA, TX)

773,401

10%

944,157

9%

1,411,635

12%

990,875

8%

688,957

6%

381,131

5%

230 - Arkla Basin

5,306

0%

3,354

0%

3,552

0%

3,551

0%

17,847

0%

12,171

0%

260 - East Texas Basin

2,434

0%

325,252

3%

48,131

0%

130

0%

1,134

0%

3,560

0%

305 - Michigan Basin

103,228

1%

159,425

2%

130,168

1%

124,802

1%

101,424

1%

73,317

1%

345 - Arkoma Basin

18,059

0%

18,152

0%

2,824

0%

6,220

0%

5,950

0%

3,614

0%

350 - South Oklahoma Folded Belt

0

0%

4,580

0%

17,422

0%

47,665

0%

38,627

0%

25,359

0%

355 - Chautauqua Platform

39,207

1%

23,253

0%

13,910

0%

9,559

0%

5,357

0%

2,692

0%

360 - Anadarko Basin

1,951,932

26%

1,079,360

10%

79,744

1%

194,986

2%

199,248

2%

8,674

0%

375 - Sedgwick Basin

0

0%

661,828

6%

0

0%

234

0%

3,033

0%

0

0%

385 - Central Kansas Uplift

71,586

1%

90,656

1%

101,570

1%

93,974

1%

28,525

0%

19,500

0%

395 - Williston Basin

3,316,405

45%

5,746,941

55%

7,863,150

67%

9,691,472

76%

8,528,583

68%

5,420,456

66%

415 - Strawn Basin

0

0%

0

0%

0

0%

6,291

0%

0

0%

0

0%

420 - Fort Worth Syncline

39,882

1%

50,428

0%

2,186

0%

4,907

0%

28

0%

25

0%

430 - Permian Basin

677,415

9%

779,460

8%

1,229,008

11%

1,051,295

8%

2,448,137

20%

1,967,988

24%

435 - Palo Duro Basin

19,829

0%

19,510

0%

3

0%

62

0%

1,866

0%

0

0%

515 - Powder River Basin

39,890

1%

77,435

1%

197,564

2%

106,907

1%

69,643

1%

41,490

1%

520 - Big Horn Basin

0

0%

0

0%

0

0%

1,088

0%

944

0%

0

0%

535 - Green River Basin

294

0%

3,626

0%

8,404

0%

3,191

0%

2

0%

0

0%

540 - Denver Basin

267,533

4%

313,901

3%

383,261

3%

228,937

2%

82,878

1%

24,899

0%

545 - North Park Basin

0

0%

0

0%

0

0%

0

0%

26,989

0%

40,151

0%

575 - Uinta Basin

22,251

0%

31,682

0%

115,014

1%

48,026

0%

34,872

0%

28,416

0%

580 - San Juan Basin

9,910

0%

10,470

0%

22,593

0%

8,536

0%

13,959

0%

13,795

0%

595 - Piceance Basin

0

0%

0

0%

0

0%

0

0%

124

0%

451

0%

745 - San Joaquin Basin

10,487

0%

3,248

0%

9,836

0%

2,557

0%

34,723

0%

7,499

0%

820 - AK Cook Inlet Basin

0

0%

0

0%

0

0%

0

0%

83

0%

1

0%

TOTAL

7,408,353

100%

10,384,996

100%

11,699,948

100%

12,709,097

100%

12,465,041

100%

8,202,349

100%

Page 18 of 20

-------
April 2018

Appendix C - GHGRP Subpart W Miscellaneous Production Flaring Emissions, by basin, for n \	x

Basin

2011

2012

2013

2014

2015

2016

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

140 - Florida Platform

0

0%

31

0%

0

0%

316

0%

95

0%

2,905

0%

160 - Appalachian Basin

0

0%

9,474

0%

156,596

3%

1,508

0%

439

0%

3,091

0%

160A - Appalachian Basin (Eastern
Overthrust Area)

10,059

0%

21,295

1%

68,263

1%

44,956

1%

51,167

1%

66,029

2%

200 - Black Warrior Basin

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

210 - Mid-Gulf Coast Basin

58,779

2%

70,807

2%

129,025

2%

150,254

3%

142,160

3%

107,848

4%

220 - Gulf Coast Basin (LA TX)

347,141

14%

625,252

16%

1,313,767

25%

1,311,265

25%

1,301,568

30%

658,331

23%

230 - Arkla Basin

2,447

0%

1,204

0%

79,635

2%

19,459

0%

24,286

1%

2,035

0%

260 - East Texas Basin

43,960

2%

25,802

1%

26,203

1%

15,192

0%

17,847

0%

12,968

0%

305 - Michigan Basin

4,949

0%

14,923

0%

4,402

0%

3,210

0%

3,230

0%

5,215

0%

345 - Arkoma Basin

13

0%

12

0%

12

0%

0

0%

24

0%

9

0%

350 - South Oklahoma Folded Belt

1,075

0%

1,552

0%

1,324

0%

2,418

0%

5,856

0%

7,982

0%

355 - Chautauqua Platform

424

0%

30,371

1%

15,108

0%

29,880

1%

3,408

0%

584

0%

360 - Anadarko Basin

142,911

6%

70,685

2%

145,823

3%

85,971

2%

232,793

5%

143,872

5%

375 - Sedgwick Basin

51

0%

0

0%

0

0%

1,394

0%

0

0%

254

0%

385 - Central Kansas Uplift

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

395 - Williston Basin

913,695

38%

488,554

12%

708,243

14%

1,473,619

28%

910,642

21%

293,829

10%

400 - Ouachita Folded Belt

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

415 - Strawn Basin

5,967

0%

5,587

0%

2,269

0%

7,491

0%

2,365

0%

0

0%

420 - Fort Worth Syncline

6,326

0%

8,690

0%

23,043

0%

35,343

1%

38,969

1%

568

0%

425 - Bend Arch

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

430 - Permian Basin

374,182

16%

2,108,306

54%

1,962,876

38%

1,508,848

29%

1,213,197

28%

1,100,472

38%

435 - Palo Duro Basin

0

0%

0

0%

0

0%

354

0%

390

0%

71

0%

450 - Las Animas Arch

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

455 - Las Vegas-Raton Basin

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

507 - Central Western Overthrust

625

0%

925

0%

701

0%

111

0%

112

0%

120

0%

515 - Powder River Basin

28,534

1%

52,245

1%

105,528

2%

125,437

2%

102,594

2%

34,839

1%

520 - Big Horn Basin

4,122

0%

2,494

0%

177

0%

1,954

0%

1,165

0%

0

0%

530 - Wind River Basin

0

0%

373

0%

528

0%

621

0%

129

0%

28

0%

535 - Green River Basin

84,576

4%

158,743

4%

255,830

5%

59,517

1%

55,234

1%

54,918

2%

540 - Denver Basin

61,760

3%

13,454

0%

10,950

0%

89,192

2%

118,692

3%

153,414

5%

545 - North Park Basin

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

575 - Uinta Basin

4,588

0%

16,025

0%

2,702

0%

12,325

0%

8,066

0%

14,383

1%

580 - San Juan Basin

394

0%

71

0%

39,238

1%

70,284

1%

28,342

1%

187

0%

585 - Paradox Basin

236,981

10%

146,578

4%

113,924

2%

161,528

3%

61,032

1%

55,460

2%

595 - Piceance Basin

14,247

1%

5,043

0%

5,828

0%

4,507

0%

3,257

0%

4,828

0%

730 - Sacramento Basin

0

0%

0

0%

0

0%

0

0%

0

0%

0

0%

Page 19 of 20

-------
April 2018

Basin

2011

2012

2013

2014

2015

2016

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

Total C02 +
CH4 Emissions
(mt C02e)

% of
Total

740 - Coastal Basins

0

0%

0

0%

0

0%

136

0%

136

0%

0

0%

745 - San Joaquin Basin

16,360

1%

13,884

0%

15,494

0%

8,547

0%

16,082

0%

26,941

1%

750 - Santa Maria Basin

0

0%

0

0%

2,204

0%

864

0%

232

0%

277

0%

760 - Los Angeles Basin

933

0%

2,486

0%

2,191

0%

1,591

0%

1,548

0%

0

0%

820 - AK Cook Inlet Basin

1,716

0%

2,118

0%

3,263

0%

2,151

0%

514

0%

490

0%

890 - Arctic Coastal Plains Province

35,172

1%

25,434

1%

26,837

1%

11,040

0%

11,188

0%

119,898

4%

TOTAL

2,401,985

100%

3,922,418

100%

5,221,983

100%

5,241,284

100%

4,356,758

100%

2,871,844

100%

Page 20 of 20

-------