April 2022

Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2020:

Updates for Post-Meter Emissions

This memorandum discusses the updates implemented in EPA's 2022 Inventory of U.S. Greenhouse Gas
Emissions and Sinks (GHGI) to include emission estimates for natural gas post-meter sources. Additional
considerations for post-meter were previously discussed in a memorandum released in September 2021
(Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2020: Updates Under Consideration for Post-
Meter Emissions).1

1	2021 (Previous) GHGI Methodology

Leak emissions beyond gas meters (e.g., such as from home heating, water heating, and stoves) and from
natural gas-fueled vehicles were not included in the GHGI. EPA included emission estimates for natural gas
post-meter sources in the current Inventory (2022 GHGI).

2	Background

Post-meter emissions are included in the 2019 Refinement to the 2006 Intergovernmental Panel on Climate
Change (IPCC) Guidelines for National Greenhouse Gas Inventories under natural gas systems (IPCC 2019
refinements). Post-meter emission sources include certain leak emissions from residential and commercial
appliances, industrial facilities and power plants, and natural gas fueled vehicles. The IPCC post-meter sub-
segments are as follows:

•	Appliances in residential and commercial sectors - Leakage from house piping and natural gas
appliances such as furnaces, water heaters, stoves and ovens, and barbecues/grills.

•	Leakage at industrial plants and power stations - Leakage from internal piping.

•	Natural gas fueled vehicles - Emissions from vehicles with alternative fuels produced from natural gas
e.g., LNG, CNG, propane. Emissions for natural gas-fueled vehicles include releases from dead volumes
during fueling, emptying of gas cylinders of high-pressure interim storage units, for execution of
pressure tests and relaxation of residual pressure from vehicles' gas tanks, or decommissioning.

3	Available Data

EPA reviewed multiple data sources to identify relevant emission factors (EF) and activity data. Each of the
data sources are identified here and a brief summary is provided.

3.1 Emission Factor Sources

EPA reviewed post-meter emissions data from the following sources:

•	IPCC 2019 - IPCC 2019 refinements contain emission factors for CH4 and C02 for post-meter emissions
from residential and commercial sector appliances, industrial plants and EGUs, and natural gas fueled
vehicles.2 The IPCC EFs for the residential and commercial sectors are appliance-based EFs (i.e.,
emissions per appliance). The industrial plants and EGUs EFs are based on consumption of natural gas
in the industrial and electricity generation sectors (i.e., emissions per volume of gas consumption).
Finally, the EFs for natural gas fueled vehicles are based on the number of natural gas vehicles (i.e.,
emissions per natural gas vehicle).

1	Stakeholder materials including draft and final memoranda for the current 1990-2020 Inventory and previous Inventories are available
at https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems.

2	2019 IPCC Refinements, Vol 2, Chap 4, Table 4.2.4k.

Page 1 of 12


-------
April 2022

•	Research Studies - EPA reviewed four research studies on post-meter emissions that were recently
published in peer-reviewed scientific journals. All the studies focused on leak emissions from
residential natural gas appliances and households.

o An Estimate of Natural Gas Methane Emissions from California Homes (Fischer et al. 2018)3 -
This study measured CH4 leak emissions from 75 homes in California that use natural gas. The
measurements captured passive house leak emissions, which included interior leaks and
quiescent appliances (with only pilot lights burning), and steady-state operation of natural gas
appliances.

o Unburned Methane Emissions From Residential Natural Gas Appliances (Merrin and Francisco
2019)4 - This study measured CH4 leak emissions from residential natural gas appliances in
Boston and Indianapolis. Natural gas space heating, water heating, and cooking appliances
were measured in over 100 homes. The methane measurements were conducted during
ignition, steady-state operation, and extinguishment phases of appliance operation.

o Quantifying Methane Emissions from Natural Gas Water Heaters (Lebel et al. 2020)5 - This
study measured CH4 leak emissions from natural gas water heaters in California. Water heaters
from 64 northern California homes were measured during ignition, steady-state operation, and
extinguishment phases of water heater operation.

o Beyond-the-Meter: Unaccounted Sources of Methane Emissions in the Natural Gas

Distribution Sector (Saint-Vincent and Pekney 2020)S-The authors performed a literature
review on residential post-meter emissions data, which included assessing some of the above
studies.

•	California Air Resource Board (CARB) GHG Inventory - California's GHG Inventory includes CH4
estimates from residential post-meter natural gas leaks. CARB uses CH4 emissions data from the
Fischer et al. 2018 study to estimate residential post-meter emissions.7

•	International GHG Inventories - EPA reviewed national GHG Inventory reports and found several that
included post-meter estimates.

o The Australian GHG Inventory uses appliance-based CH4 EFs from the Merrin and Francisco
2019 study to estimate residential appliance post-meter emissions.

o The German GHG Inventory includes leak estimates from natural gas meters and fittings in the
residential, commercial, and industrial sectors, and leak estimates from natural gas fueled
vehicles. The CH4 EFs used by Germany are based on local data and studies.

o The UK GHG inventory includes residential and commercial leak emissions that are based on
country-specific EFs for natural gas consumption (i.e., kg CH4/TJ).

3.2 Activity Data Sources

EPA reviewed the following sources of activity data:

3	Marc L. Fischer, Wanyu R. Chan, Woody Delp, Seongeun Jeong, Vi Rapp, Zhimin Zhu. An Estimate of Natural Gas Methane Emissions
from California Homes. Environmental Science & Technology 2018, 52 (17), 10205-10213.
https://pubs.acs.ore/doi/10.1021/acs.est.8b03217.

4	Zachary Merrin, Paul W. Francisco. Unburned Methane Emissions from Residential Natural Gas Appliances. Environmental Science &
Technology 2019, 53 (9), 5473-5482. https://doi.org/10.1021/acs.est.8b05323.

5	Eric D. Lebel, Harmony S. Lu, Simone A. Speizer, Colin J. Finnegan, Robert B. Jackson. Quantifying Methane Emissions from Natural Gas
Water Heaters. Environmental Science & Technology 2020, 54 (9), 5737-5745. https://doi.ore/10.1021/acs.est.9b07189.

6	Patricia M. B. Saint-Vincent, Natalie J. Pekney. Beyond-the-Meter: Unaccounted Sources of Methane Emissions in the Natural Gas
Distribution Sector. Environmental Science & Technology 2020, 54 (1), 39-49. https://doi.org/10.1021/acs.est.9b04657.

7	CARB GHG Inventory Updates Documentation, 2019 Edition. Available online at:

https://ww3.arb.ca.eov/cc/inventorv/pubs/reports/2000 2017/ehe inventory 00-17 method update document.pdf.

Page 2 of 12


-------
April 2022

•	Residential sector - Data available to estimate post-meter emissions from residential natural gas
appliances are national counts of natural gas households. National data on housing counts, by fuel
type and by specific end uses such as heating and cooking, are available from the U.S. Census Bureau's
American Housing Survey (AHS) and Energy Information Administration's Residential Energy
Consumption Survey (RECS).8,9 Both the datasets include national counts on number of housing units
using natural gas for specific end uses such as space heating, water heating, cooking, clothes drying,
pools and spa heating, and outdoor grills. The data do not include natural gas appliance counts.

o Housing Counts:

¦	AHS. National data on housing counts, by fuel type, are available from the U.S. Census
Bureau's AHS.10 National summary data are reported biennially by AHS starting with
1973.

¦	RECS. In addition to the U.S. Census Bureau, national data on housing counts by fuel
type are also available from the Energy Information Administration (EIA). The EIA
periodically conducts the nationwide Residential Energy Consumption Survey (RECS).
The RECS data are only available for 7 years in the 1990-2020 time-series (i.e., 1990,
1993, 1997, 2001, 2005, 2006, and 2015).

•	Commercial sector - Data available to estimate post-meter emissions from commercial natural gas
appliances are national data on commercial buildings, by fuel types and end uses such as heating and
cooking from ElA's Commercial Buildings Energy Consumption Survey (CBECS)nand commercial meter
counts available from EIA. Natural gas appliance counts are unavailable.

o Commercial building counts: The CBECS contains national data on commercial buildings by
type of fuel and energy end use (i.e., space heating, water heating, and cooking). The CBECS
does not contain information on number of natural gas appliances used in commercial
buildings; however, the survey indicates the number of commercial buildings that use natural
gas for a particular end use. The CBECS is only available for 1992, 1995, 1999, 2003, and 2012
(i.e., 5 years in the 1990-2020 time-series). Data for 2018 CBECS are not publicly released yet.

o Commercial meter counts: National commercial appliance counts can be estimated by

multiplying a default value of appliances per commercial natural gas meter by the number of
commercial natural gas meters from EIA.12 Time-series data for commercial natural gas meters
from EIA is used in the GHGI for estimating emissions from meters in the distribution segment.

•	Industrial plants and EGUs - Annual national natural gas consumption data for industrial and electricity
generation sectors are available from the Monthly Energy Review publications of the EIA.13

•	Natural gas fueled vehicles - National vehicle population estimates are available from EPA's Motor
Vehicle Emission Simulator model (MOVES).14 The latest version of the MOVES model, MOVES3,
contains natural gas vehicle fleet population estimates for 1990 and 1999-2020 and incorporates
updated data on vehicle population, travel activity, and fuel supply information. Additionally, ElA's

8	U.S. Census Bureau's American Housing Survey (AHS). https://www.census.gov/programs-surveys/ahs.html.

9	Energy Information Administration's Residential Energy Consumption Survey (RECS). https://www.eia.gov/consumption/residential/.

10	The U.S. Census also publishes the American Community Survey (ACS), which contains national housing counts by type of house
heating fuel. Unlike the AHS, ACS data are only available for natural gas space heating end use and don't include other end uses such as
water heating, cooking, and clothes drying, and ACS data are only available from 2010 forward. EPA did not consider this data source
for the update under consideration.

11	ElA's Commercial Buildings Energy Consumption Survey - https://www.eia.eov/consumption/commercial/.

12	U.S. Energy Information Administration. Number of Natural Gas Consumers.
https://www.eia.gov/dnav/ng/ng_cons_num_dcu_nus_a.htm.

13	U.S. Energy Information Administration. April 2021 Monthly Energy Review.
https://www.eia.gov/totalenerev/data/monthlv/archive/00352104.pdf.

14	U.S. EPA's Latest MOVES Model: https://www.epa.gov/moves/latest-version-motor-vehicle-emission-simulator-moves.

Page 3 of 12


-------
April 2022

Alternate Fuel Data Center (AFDC) contains the Alternate Fuel Vehicle Inventory.15 This inventory
contains natural gas fueled vehicle counts for 2004-2019 for cities participating in the U.S. Department
of Energy's Clean Cities Coalition Network.16

4 Analysis of Available Data

This section presents the available emission factor data and activity data for the post-meter sub-segments -
Appliances in residential and commercial sectors; Leakage at industrial plants and power stations; and Natural
gas fueled vehicles.

4.1 Emission Factors

The following sections summarize EFs from various sources reviewed by EPA and identify the EFs used in the
update for each sector.

4.1.1 Residential Sector

Residential sector CH4 EFs for natural gas leak emissions from house piping and appliances are shown in Table
1. All the residential sector EFs presented in Table 1 include leaks from natural gas appliances (e.g., during
steady state operations); however, only a single source (Fischer et al. 2018) also accounts for passive leaks
from residential natural gas piping (i.e., quiescent house leakage).

Table 1. CH4 Emission Factors - Post-Meter Leaks in Residential Sector.

Data Source

ch4ef

EF Units

EF Includes
Appliance
Leaks?

EF Includes Passive
House Leaks?

IPCC

4.0

Kg/appliance

Yes

Unspecified

CARB/Fischer etal. (2018)

2.54

Kg/NG House

Yes

Yes

Merrin and Francisco (2019)

0.43

Kg/NG House

Yes

No

Lebel et al. (2020)

1.42

Kg/NG Water
Heater

Yes

No

2021 Australia GHGI (1990-
2019)

0.06-1.2

Appliance based3

Yes

No

2021 UK GHGI (1990-2019)

1.9

Kg/TJ natural gas

Yes

No

aThe EFs were derived from the Merrin and Francisco 2019 study (CH4 EFs for ovens, stoves, furnaces, and water heaters). The
appliance-based emission factors range from 0.06 kg/stove to 1.2 kg/tankless water heater.

Another study (Saint-Vincent and Pekney, 2020), not listed in Table 1 above, estimated national emissions
using emissions data from other studies. This study combined passive house leak emissions data from the
Fischer et al. study and appliance emissions from the Merrin and Francisco study. The implied CH4 EF
developed by this study is 2.41 kg/natural gas household (i.e., national emissions estimated by study divided
by national number of households using natural gas).

The California Air Resources Board (CARB) GHG Inventory developed an estimate for post-meter emissions
using the Fischer et al. study. The EF calculated from the Fischer et al. study accounts for passive house leak
emissions and appliance leak emissions.

15	Clean Cities Alternate Fuel Vehicle Inventory: https://afdc.enerev.gov/data/10581.

16	Clean Cities Coalition Network: https://cleancities.energy.gov/coalitions/.

Page 4 of 12


-------
April 2022

The Australian GHGI (1990-2019) uses residential appliance CH4 EFs from the Merrin and Francisco study to
estimate residential post-meter emissions.17

The UK GHGI (1990-2019) estimates post-meter emissions in the residential and commercial sectors.18
Leakages are estimated for a range of different appliances that use gas, combined with national statistics on
natural gas consumption in the domestic and commercial sectors. The UK GHGI includes appliances used for
space heating, water heating, and cooking in the residential sector and appliances (ovens and boilers) used in
commercial catering and other service sectors.

IPCC 2019 also provided a residential sector post-meter C02 EF of 0.033 kg/appliance. None of the other data
sources provided C02 emissions data.

In the final 2022 GHGI methodology, EPA used the CH4 EF from the Fischer et al. 2018 study to estimate
emissions from residential post-meter sources. EPA deducted CH4 emissions from the residential natural gas
fuel combustion source (Fossil Fuel Combustion source in GHGI)19 from the Fischer-derived estimate to avoid
double-counting. EPA did not include C02 estimates for residential post-meter sources in the final 2022 GHGI
methodology as these are captured under residential natural gas fuel combustion in the GHGI.

4.1.2 Commercial Sector

Commercial sector CH4 EFs for natural gas appliance leak emissions are shown in Table 2.

Table 2. Emission Factors - Post-Meter Leaks in Commercial Sector.

Data Source

ch4ef

C02 EF

EF Units

IPCC

4.0

0.033

Kg/appliance

EPA used the IPCC CH4 and C02 EFs (Table 2) to estimate emissions from commercial post-meter sources in the
2022 GHGI.

4.1.3 Industrial Plants and EGUs

CH4 EFs for natural gas leak emissions at industrial plants and EGUs are shown in Table 3.

Table 3. Emission Factors - Post-Meter Leaks from Industrial Plants and EGUs.

Data Source

CH4 EFa

EF Units

IPCC

11,326.7

Kg/billion cubic feet gas consumed

2021 Germany GHGI (1990-2019)

7,702.2b

Kg/billion cubic feet gas consumed

a Converted CH4 EFs from kg/million cubic meters gas consumed to kg/billion cubic feet gas consumed
using a conversion factor of 35.3147 cubic feet/cubic meter.

bThe original CH4 EF from the German National Inventory Report (NIR) is 0.4 m3 CH4/1,000 m3 natural gas
consumption. The conversion assumed density of CH4 to be 19.26 g/ft3 (0.68 kg/m3).

IPCC 2019 also provided a C02 EF of 3.3 Kg/million cubic meters gas consumed for post-meter leaks at
industrial plants and EGUs. EPA used the IPCC CH4 and C02 EFs to estimate emissions from industrial post-
meter sources in the 2022 GHGI.

17	Australia. 2021 National Inventory Report (1990-2019), Volume 1, Table 3.49b. April 2021. Available online at:
https://unfccc.int/documents/273478.

18	United Kingdom. 2021 National Inventory Report (1990-2019), Section 3, MS 19. April 2021. Available online at:
https://unfccc.int/documents/273439.

19	https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks.

Page 5 of 12


-------
April 2022

4.1.4 Natural Gas Fueled Vehicles

IPCC 2019 default EFs for CH4 and C02 are shown in Table 4. The IPCC EFs were derived from a German study
and are also used in the German GHGI.

Table 4. Emission Factors - Post-Meter Leaks from Natural Gas Fueled Vehicles.

Data Source

ch4ef

C02 EF

EF Units

IPCC

0.33

0.0023

Kg/vehicle

EPA used the IPCC CH4 and C02 EFs (Table 4) to estimate post-meter emissions from natural gas fueled vehicles
in the 2022 GHGI.

4.2 Activity Data
4.2.1 Residential Sector

Activity data available for the residential sector post-meter estimates are the national number of households
using natural gas as fuel. Table 5 contains a summary of available RECS and AHS housing units data (both total
houses and houses using natural gas as a fuel) for the seven years for which RECS data are available and the
most recent year (2019) that AHS data are available. National data in Table 5 represent occupied housing units.

Table 5. Summary of National Housing Data (millions of housing units).

Year

EIA/RECS

Census/AHS

% Difference NG
Housing Units

Total Housing Units

NG Housing Units

Total Housing Units

NG Housing Units

1990

94.0

57.7

Not Available

Not Available

-

1993

96.6

58.4

94.7

65.6

-0.3%

1997

101.5

61.9

99.5

60.5

-2%

2001

107.0

66.9

106.3

83.1

-3%

2005

111.1

69.5

108.9

86.6

-4%

2009

113.6

69.2

111.8

86.2

-2%

2015

118.2

68.6

118.3

80.3

17%

2019

Not Available

Not Available

124.1

84.7

-

The AHS and RECS datasets also contain the number of housing units using natural gas for specific end use
purposes such as space heating, water heating, cooking, clothes drying, outdoor grilling, and spa/pool heating.
Natural gas appliance counts are not available in the AHS and RECS data. In Table 6 below, national appliance
counts were estimated by assuming one appliance of each type per housing unit using natural gas for that
appliance type. For example, if the data indicate 60,000 housing units that use natural gas for water heating,
then the number of natural gas water heaters is assumed to be 60,000. Table 6 contains a summary of
residential natural gas appliance counts from RECS and AHS for the seven years for which RECS data are
available and the most recent year (2019) that AHS data are available.

Table 6. Summary of Estimated National Residential Appliance Data (millions of units).

Year

Based on
EIA/RECS

Based on
Census/AHS

1990

157

NA

1993

156

149

1997

164

159

2001

179

170

2005

185

176

2009

185

174

Page 6 of 12


-------
April 2022

Year

Based on
EIA/RECS

Based on
Census/AHS

2015

189

176

2019

NA

181

EPA used the national count of natural gas households from the AHS data for the final 2022 GHGI methodology
for residential post-meter sources.

4.2.2 Commercial Sector

ElA's CBECS contains data on the number of commercial buildings that use natural gas for specific end uses
such as space heating, water heating, and cooking but does not indicate the number of appliances at
commercial buildings. The commercial buildings included in the CBECS are categorized by their primary
business activity. The commercial building categories covered by the CBECS include education, food service,
health care, lodging, mercantile, offices, religious worship, public services, and others (e.g., laboratories,
manufacturing with retail space, data centers, and crematoriums).

The CBECS data are only available for 1992, 1995, 1999, 2003, and 2012. Data from the 2018 CBECS are yet to
be published.20 Table 7 presents the national summary for the available years of CBECS data. The CBECS data
do not contain commercial appliance counts. National appliance counts can be estimated by assuming one
appliance of each type per commercial building using natural gas for that appliance type. For example, if the
data indicate 100,000 commercial building use natural gas for water heating, then the number of natural gas
water heaters is assumed to be 100,000. The estimated natural gas appliance counts are also shown in Table 7,
below. In addition, the number of commercial natural gas meters are also shown in Table 7 for the years CBECS
data are available. Using the derived appliance counts and the commercial meter counts, EPA developed
estimates for appliances per commercial meter.

Table 7. Summary of CBECS Data (1000s).



1992

1995

1999

2003

2012

All commercial buildings

4,806

4,579

4,657

4,645

5,557

Commercial buildings using natural gas as fuel for t

ie following enc

uses:

Primary space-heating energy source

2,276

2,106

2,189

1,999

2,322

Water-heating energy source

1,647

1,577

1,520

1,445

1,758

Cooking energy source

431

448

505

457

740

Cooling energy source

106

65

142

17

12

Estimated Appliance Count

4,460

4,196

4,356

3,918

4,832

Commercial Meter Count - 2022 GHGI

4,409

4,636

5,010

5,152

5,356

Estimated Appliances/Commercial Meter3

1

1

1

1

1

a All values rounded to the closest integer. The actual values range from 0.76 to 1.01 appliances/meter.

To estimate the number of commercial appliances, EPA used the number of commercial natural gas meters
from EIA and the estimate of one appliance per commercial meter for the final 2022 GHGI methodology for
commercial post-meter sources.

4.2.3 Industrial Plants and EGUs

Activity data used to estimate post-meter leak emissions from industrial plants and EGUs are national natural
gas consumption for the industrial and electric power sectors. EIA provides national data on consumption of
natural gas by the industrial and electric power sectors for the entire time-series (1990-2020); see Table 8.

20 Preliminary 2018 CBECS data are not yet published. Website indicates detailed tables will be available in spring/summer 2021.

Page 7 of 12


-------
April 2022

Table 8. Summary of Natural Gas Consumption By Industrial and Electric Power Sectors (BCF).

Year

Industrial Sector

Electric Power Sector

1990

8,255

3,245

1991

8,360

3,316

1992

8,698

3,448

1993

8,872

3,473

1994

8,913

3,903

1995

9,384

4,237

1996

9,685

3,807

1997

9,714

4,065

1998

9,493

4,588

1999

9,158

4,820

2000

9,293

5,206

2001

8,463

5,342

2002

8,640

5,672

2003

8,273

5,135

2004

8,354

5,464

2005

7,713

5,869

2006

7,669

6,222

2007

7,881

6,841

2008

7,890

6,668

2009

7,443

6,873

2010

8,112

7,387

2011

8,317

7,574

2012

8,622

9,111

2013

8,909

8,191

2014

9,158

8,146

2015

9,098

9,613

2016

9,274

9,985

2017

9,533

9,266

2018

10,112

10,590

2019

10,268

11,288

2020

10,086

11,616

EPA used EIA natural gas consumption data in the final 2022 GHGI methodology for industrial and EGU post-
meter sources.

4.2.4 Natural Gas Fueled Vehicles

Activity data required to estimate post-meter leak emissions for natural gas fueled vehicles are the national
population of natural gas fueled vehicles. EPA evaluated data from MOVES3 and AFDC. MOVES3 includes the
annual compressed natural gas (CNG) vehicle population for 1990 and 1999 through 2020. The population of
vehicles running on liquefied natural gas (LNG) or renewable natural gas (RNG) are not available in MOVES3.
CNG vehicle counts from MOVES3 include buses (transit and school buses), refuse trucks, single-unit trucks,
and combination trucks (i.e., heavy-duty vehicles). Currently, there is only a single OEM that offers a factory

Page 8 of 12


-------
April 2022

built CNG light-duty vehicle in the U.S. market.21 However, there are options for after-market CNG conversions
for passenger automobiles. Converted passenger automobile counts are not included in MOVES3.22

The AFDC provides an annual alternate fuel vehicle inventory for 2004-2019 compiled from the Clean Cities
Coalition. The AFDC data is not available at the national-level. The alternate fuel vehicle inventory only
includes the cities that are part of the Clean Cities Coalition. An estimated 79 percent of the national
population lives inside the boundaries of the Clean Cities Coalition Network. Table 9 summarizes the MOVES3
and AFDC data.

Table 9. Summary of Vehicle Populations from MOVES3 and Clean Cities Alternate Fuel Vehicle

Inventory.

Year

MOVES3
CNG Vehicles

Clean Cities Alternate Fuel Vehicle Inventory

CNG Vehicles

LNG Vehicles

RNG Vehicles

All Natural Gas Vehicles

1990

6

NA

NA

NA

NA

1999

10,139

NA

NA

NA

NA

2000

13,063

NA

NA

NA

NA

2001

15,787

NA

NA

NA

NA

2002

19,229

NA

NA

NA

NA

2003

20,963

NA

NA

NA

NA

2004

22,364

76,257

0

0

76,257

2005

23,695

49,271

1,873

0

51,144

2006

24,093

57,458

2,271

0

59,729

2007

25,028

55,021

1,731

0

56,752

2008

26,959

51,121

2,053

0

53,174

2009

30,289

44,317

2,038

0

46,355

2010

30,708

42,911

3,410

0

46,321

2011

33,557

48,157

4,315

0

52,472

2012

37,925

59,521

3,411

0

62,932

2013

42,311

79,616

3,645

0

83,261

2014

50,317

68,479

2,992

313

71,784

2015

60,132

107,283

3,974

366

111,623

2016

69,897

98,388

4,924

1,157

104,469

2017

80,021

97,271

5,070

1,734

104,075

2018

89,033

82,266

5,100

1,677

89,043

2019

98,535

100,938

4,917

4,922

110,777

2020

107,519

NA

NA

NA

NA

NA = not available

EPA used CNG vehicle counts from MOVES3 for the final 2022 GHGI methodology for natural gas vehicle post-
meter sources.

5 Time Series Considerations

•	Emission Factors: Currently, there is no time series information on emission factors available for post-
meter emissions. The emission factors in the update were held constant over the time series.

•	Residential Activity Data:

21	https://ngvamerica.org/vehicles/.

22	MOVES population and activity technical report is available online at: https://nepis.epa.eov/Exe/ZvPDF.cei?Dockev=P1011TF8.pdf.

Page 9 of 12


-------
April 2022

o U.S. Census Bureau's American Housing Survey (AHS) - National housing data by type of fuel
and fuel end use purpose are available from AHS. National summary data from the AHS are
published biennially (published every odd year). For the 2022 GHGI, natural gas households
counts for missing years were gap filled as the average value of the two most recent years for
which data are available (i.e., years immediately before and after the missing year).

•	Commercial Activity Data - EPA estimated the number of appliances per commercial meter using data
from the CBECS and the commercial meter counts data from EIA (see Table 7). The data indicate 1
appliance/commercial meter for all years with CBECS data. In the final 2022 GHGI methodology, EPA
applied the same value (1 appliance/commercial meter) for all time-series years.

•	Industrial and Power Plants Activity Data -ElA's natural gas consumption data for the industrial and
electric power sectors are available for the entire time series.

•	Natural Gas Vehicles Activity Data - MOVES3 National Vehicle Population - National vehicle counts are
available from MOVES3 for 1990 and 1999-2020. In the final 2022 GHGI methodology, EPA used ElA's
data on annual natural gas vehicular fuel consumption for 1990-1999 to gap fill MOVES3 vehicle counts
for 1991-1998. EPA calculated the ratio of natural gas vehicle fuel consumption in each of the missing
years (i.e., 1991-1998) to natural gas vehicle fuel consumption in 1999. The 1999 vehicle count from
MOVES3 was then multiplied by the year specific ratio to gap fill MOVES3 vehicle counts for 1991-
1998.

6 National Emissions Estimates for Post-Meter in the 2022 GHGI

Table 10 presents a summary of the 2020 national estimates for post-meter sources developed using the final
2022 GHGI methodology. The EFs and activity data incorporated into the final 2022 GHGI methodology are
presented in sections 3 and 4, above.

Table 10. 2020 National Estimates From Post-Meter Sources (2022 GHGI).

Post-Meter
Source

Activity

ch4ef

COzEF

2020 Emissions (2022 GHGI)

CH4 (metric
tons)

C02 (metric
tons)

Residential

84,726,000 NG
houses

2.54 kg/NG House

NA

192,199a

NA

Commercial

5,626,925
appliances

4 kg/appliance

0.033

kg/appliance

22,508

186

Industrial &
EGUs

21,571 Bcf

11,326.7 kg/bcf

3.3 kg/million
cubic meters

244,333

2,016

Natural Gas
Vehicles

107,519 vehicles

0.33 kg/vehicle

0.0023
kg/vehicle

32

0.2

Total







459,072

2,202

a. Incorporates a deduction of 23,005 mt Cm for residential natural gas combustion.

7 Requests for Stakeholder Feedback

EPA sought stakeholder feedback on the approaches under consideration through two 2021 webinars, in the
September 2021 memo, and in the public review draft of the GHGI. EPA received comments on the September
2021 version of the Post-Meter Memo and through the public review draft of the Inventory. These
comments included recommendations to delay inclusion of post-meter estimates to the GHGI until further
research studies have been conducted. The comments cautioned against the use of residential post-meter EFs
developed from a research study conducted in California, asserting that it is not representative at the national-

Page 10 of 12


-------
April 2022

level. EPA also received comment on appliance pilot light phase-out and potential implications on time-series
estimates.

The questions below were not updated for this memorandum and are copied from the September 2021
memo.

1.	EPA seeks additional data sources to be considered for post-meter emissions (emission factor and
activity data sources).

2.	EPA seeks stakeholder feedback on time series data for post-meter emission sources. Are data
available that would allow the GHGI to reflect changes over time in average emissions per emission
source for post-meter?

Page 11 of 12


-------
April 2022

Appendix A - Measurement Methodologies from Data Sources Considered for Updates

Emission
Source

Measurement and/or Calculation Type

# Sources

Location &
Representativeness

EF Calculation Method

An Estimate of Natural Gas Methane Emissions from California Homes (Fischer et al. 2018)

Quiescent

whole-house

emissions

Appliance
emissions

Mass balance approach - Controlled flow of outdoor air was used to
ventilate the house, while measuring both the indoor and outdoor
air CH4 concentrations over time. Once indoor CH4 concentration
reached steady state, the enhancement of indoor CH4 relative to
outdoor air combined with the known volumetric flow rate of air
was used to estimate indoor CH4 emissions. CH4 was measured with
a portable gas analyzer.

Methane emissions were measured during steady operation for two
NG appliances in each home. Measurements were made using the
same portable gas analyzer used for whole-house measurements.
Each appliance was operated for 10-15 min before the
measurement.

75 owner-occupied, single-
family detached homes that
use NG for at least two of the
following purposes: space
heating, water heating,
cooking, and clothes drying.

30 homes were
located in Northern
California and
Central Valley and
45 homes were
located in Southern
California.

Study estimated state-level emissions
from whole-house and NG appliances
using measurement results and
Bayesian Markov chain Monte Carlo
sampling combined with California
housing statistics and gas use
information. EFs were derived by
using state-level estimates from the
study and the number of natural gas
homes in CA.

Unburned Methane Emissions From Residential Natural Gas Appliances (Merrin and Francisco 2019)

Appliance
emissions

Gas concentrations were measured during appliance ignition,
operation, extinguishment, and cool down. Appliances were tested
using a Picarro portable gas concentration analyzer (ppm of CH4,
C02, and water vapor; sample temperature and pressure). Cooktop
burners were tested using "CO Hot Pot" (device mimics cooking
vessel and has sampling port for measurements).

Space heating, water heating,
and cooking appliances were
tested (furnaces, boilers,
stoves, ovens, water heaters,
outdoor grills, and space
heaters from 100 residential
sites in MA, IN, IL, and NY
were tested.

72 sites in Boston,
MA and

Indianapolis, IN. 28
Additional sites in IL
and NY.

Annual CH4 emissions from an
appliance type were calculated by
combining average measured
concentration, calculated exhaust
flow, and appliance usage
assumptions (days used per year,
activations per day, average
operational duration in mins, etc.).

Quantifying Methane Emissions from Natural Gas Water Heaters (Lebel et al. 2020

NG water
heaters

Developed a high-flow sampling system to capture and measure the
emissions from the water heaters. A local background measurement
was taken 30 or more feet upwind from the WH exhaust each hour.
The concentrations of CH4 and C02 were measured using a Cavity
Ring-Down Spectrometer G2210-i (Picarro Inc.).

Emissions were measured: (1) before the appliance was turned on;
(2) as the appliance was turning on, producing a pulse of CH4; (3)
until the concentration profile reached a steady concentration to
measure incomplete combustion for a minimum of 2 min and
multiple replicates; and (4) as the water heater turned off which
typically created a second pulse of CH4.	

Emissions data were collected
from 35 residential water
heaters. Usage data were
collected from 46 residential
water heaters.

64 single-family
homes with natural
gas water heaters
in Northern
California. Water
heaters included
both storage and
tankless type.

The measured emissions data were
combined with the usage data to
estimate annual emissions.

Page 12 of 12


-------