September 2020
Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2019:
Updates Under Consideration for Natural Gas Customer Meter Emissions
This memorandum discusses updates under consideration for the 2021 U.S. Inventory of U.S. Greenhouse Gas
Emissions and Sinks (GHGI) for industrial and commercial meters.
1	Current GHGI Methodology
EPA most recently updated the GHGI emissions calculation methodology for industrial and commercial meters
in the 2016 GHGI by incorporating findings from a Gas Technology Institute (GTI) 2009 study1 to estimate
emissions. EPA's April 2016 memo Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2014: Revisions
to Natural Gas Distribution Emissions2 documents the historical considerations and the full methodology used
for industrial and commercial meters in the current GHGI.
In the current GHGI, EPA estimates industrial and commercial meter emissions using Energy Information
Administration (EIA) meter counts in each year paired with the GTI 2009 study commercial meter emission
factor (EF) of 9.7 kg/meter/yr for both commercial and industrial meter types.3 EPA applied the commercial
meter EF to both commercial and industrial meters due to the limitations of available industrial meters data
for revising EFs and based on stakeholder feedback.
2	Available Data
Two available data sources, 2009 and 2019 studies from GTI, include emissions data for industrial and
commercial meters.
The current emission factors in the GHGI come from a 2009 report by GTI and Innovative Environmental
Solutions for Operations Technology Development (OTD) that investigated methane emission factors for select
distribution sources (GTI 2009).1 The emission sources included both metering and regulating (M&R) stations
and customer meters. The GTI 2009 study conducted sampling of customer meters using screening and Hi-
Flow Samplers to quantify total emissions from leaks and vents. The GTI 2009 study sampled 836 commercial
meters at six companies and 46 industrial meters at five companies in five geographical regions across the
United States. The study included both leak and vented emissions. An average EF was determined for each
company and an overall average EF was then calculated based on the number of meters tested for each
company.
1	Gas Technology Institute and Innovative Environmental Solutions, Field Measurement Program to Improve Uncertainties for Key
Greenhouse Gas Emission Factors for Distribution Sources, November 2009. GTI Project Number 20497. OTD Project Number 7.7.b.
2	Available at 
3	EIA defines the industrial sector as, "An energy-consuming sector that consists of all facilities and equipment used for producing,
processing, or assembling goods... Overall energy use in this sector is largely for process heat and cooling and powering machinery, with
lesser amounts used for facility heating, air conditioning, and lighting. Fossil fuels are also used as raw material inputs to manufactured
products. This sector includes generators that produce electricity and/or useful thermal output primarily to support the above-
mentioned industrial activities." EIA considers the commercial sector to include "service-providing facilities and equipment of
businesses; Federal, State, and local governments; and other private and public organizations, such as religious, social, or fraternal
groups. The commercial sector includes institutional living quarters. It also includes sewage treatment facilities. Common uses of
energy associated with this sector include space heating, water heating, air conditioning, lighting, refrigeration, cooking, and running a
wide variety of other equipment. Note: This sector includes generators that produce electricity and/or useful thermal output primarily
to support the activities of the above-mentioned commercial establishments." https://www.eia.eov/tools/elossarv/index.php?id=l
Page 1 of 10

-------
September 2020
The 2019 GTI study4 conducted sampling of customer meter sets using a combustible gas indicator (CGI) to
screen for leaks and Hi-Flow Samplers to measure leak rates. Meter sets are defined as "the meter plus all
components associated with that meter up to the point of transfer of hardware responsibility to the [gas]
customer." This would include valves, flanges, tees, and other additional components associated with a meter.
All components at a meter set were first scanned with the CGI to locate all leak indications with a
concentration of at least 100 ppm. Depending on the campaign, leaks were then determined to be quantifiable
when above either 22,500 ppm or 100 ppm. To quantify leak rates at the lower concentration threshold of 100
ppm, an ultraportable greenhouse gas analyzer (UGGA) was incorporated into the Hi-Flow Sampler to increase
its measurement sensitivity. For eight of the thirteen campaigns, only leaks with indications above 22,500 ppm
were quantified. For the remaining five campaigns, leaks above 100 ppm were quantified. Leaks that did not
meet the concentration threshold of the campaign were not measured and quantified. Meters with a
quantifiable leak were then measured using Hi-Flow Samplers. Unlike the 2009 study, the 2019 study only
screened for leak emissions; vented emissions from regulators, pneumatic devices, or other sources were not
included unless they were malfunctioning. An example of a malfunction included in the 2019 study would be a
valve stuck in a position different from the intended vent position. The study sampled 186 meters at industrial
locations and 337 meters at commercial sites in six geographic regions across the United States. An average EF
was determined for each region as well as for the total population of commercial or industrial meters. The
2019 study also estimated leaker-only EFs and EFs specific to each meter type (rotary, diaphragm, etc.).
Appendix B provides an overview of both the 2009 and 2019 study designs.
The GTI 2019 commercial meter EF (leaks only) is six times higher than the GTI 2009 study (which included
both leaks and venting emissions). The 2009 value was quantified from a survey of 836 commercial meters,
while the 2019 study surveyed 337 meters. GTI 2009 quantified an EF for industrial meters leaks and venting
emissions that is higher than what was quantified by GTI 2019 for only leak emissions. The 2009 study
surveyed 46 industrial meters, while the 2019 study surveyed 186. The small sample size of the GTI 2009 study
and the wide variation observed among industrial meters led EPA to use the commercial EF for both
commercial and industrial meter emissions estimates in 2016.
3 Analysis of Available Data
This section summarizes EPA's analyses of the recently published GTI 2019 study and considerations toward
using data from either or both the 2009 and 2019 GTI studies to update the customer meters methodology in
the 2021 GHGI.
1 1 ! 1 tudy
GTI 2019 presented multiple approaches for calculation of national emissions for potential use in the GHGI.
Each approach is discussed in the following sections.
3.1.1 Population Emission Factors
GTI 2019 measured emissions rates at commercial and industrial meters in six regions across the country and
calculated population EFs from the complete population of data. Meters that did not have a quantifiable leak,
at the threshold used for each measurement campaign, were considered to have zero emissions in the
population EF calculation. GTI found that adding the non-quantified leaks (e.g., leaks with a concentration less
than 22,500 ppm but the particular measurement campaign was only quantifying leaks greater than 22,500
ppm) had a negligible impact on the mean population emission rate. Of the meter sets sampled, 82 percent of
commercial meters and 87 percent of industrial meters were found to be leaking. Of those meter sets with
4 Gas Technology Institute and US Department of Energy, Classification of Methane Emissions from Industrial Meters, Vintage vs
Modern Plastic Pipe, and Plastic-lined Steel and Cast-iron Pipe. June 2019. GTI Project Number 22070. DOE project Number ED-
FE0029061.
Page 2 of 10

-------
September 2020
leaks, 53 percent of commercial meters with leaks and 49 percent of industrial meters with leaks had
quantifiable leaks. In the 2019 study, GTI recommends EPA use separate EFs for commercial and industrial
meters. Table 1 presents the study's population EFs and sampling data.
Table 1. GTI 2019 Commercial and Industrial Meter Population CH4 EFs and Sampling Data
Parameter
Commercial Meters
Industrial Meters
EF (kg/meter/yr)
57.4
117.8
# Meter Sets Sampled
337
186
# Meter Sets with Leak Indication
278
161
# Meter Sets with Quantifiable Leak
146
79
3.1.2 Regional Emission Factors
GTI 2019 conducted regional sampling of both commercial and industrial meters and found regional variation
in EFs for both. Table 2 below shows the number of meter sets sampled and calculated EFs for each region. GTI
recommends that EPA use regional EFs separated by commercial and industrial meter types. GTI stated that
regional variation in EFs is in part due to differences in the main meter set type commonly used in a region. For
example, in the Southeast region 75 percent of meters measured were turbine meters (which GTI found to
have the highest emissions). Additionally, GTI noted it is possible that differences in leak identification and
repair procedures in each region explain the variation. Their analysis found that the regions with the highest
EFs also had the highest likelihood of finding a large leak.
Table 2. GTI 2019 Commercial and Industrial Meter Regional CH4 EFs (kg/meter/yr) and Sampling Data
Region
Commercial Meter
Sets Sampled
Commercial
Meter EF
Industrial Meter
Sets Sampled
Industrial
Meter EF
Midwest
99
28.4
77
52.3
Northeast
75
20
13
172.5
Pacific
63
4
52
17.4
Rocky Mountain
12
108.4
9
322.5
Southeast
5
139.3
15
291.7
Southwest
83
153.9
20
372.9
All
337
57.4
186
117.8
3.1.3 Leaker Emission Factors
In the 2019 study, GTI found that 43% of the meter sets sampled had a quantifiable leak. GTI's analysis found
that the Pacific region had the highest likelihood of having no leaks or small leaks, and that the Rocky
Mountain, Southeast, and Southwest regions had the highest likelihood of finding substantial leaks. This
difference is reflected in the leaker EFs for these regions. Table 3 shows the leaker EFs by region.
Table 3. GTI 2019 Commercial and Industrial Meter Regional Leaker CH4 EFs (kg/meter/yr) and
Sampling Data
Region
Commercial Meter Sets
with Quantifiable Leak
Commercial
Meter Leaker EF
Industrial Meter Sets
with Quantifiable Leak
Industrial Meter
Leaker EF
Midwest
58
48.5
35
260.0
Northeast
20
75.1
6
564.9
Pacific
28
9.0
9
233.3
Page 3 of 10

-------
September 2020
Region
Commercial Meter Sets
with Quantifiable Leak
Commercial
Meter Leaker EF
Industrial Meter Sets
with Quantifiable Leak
Industrial Meter
Leaker EF
Rocky Mountain
4
325.3
5
745.9
Southeast
4
174.1
15
707.1
Southwest
32
399.1
9
1045.8
All
146
132.4
79
277.4
3.2 Combined	taset
In addition to considering options that use the 2019 GTI dataset alone to update the GHGI, EPA is considering
options to combine the results from both the 2009 and 2019 studies to develop weighted average population
EFs. EPA used the number of samples in the respective studies to weight the EFs.
When considering both datasets, EPA evaluated leak versus vented emissions; the GTI 2009 study measured
leak and vented emissions and the GTI 2019 study focused on leak emissions only. The inclusion of vented
emissions leads to significantly higher industrial meter EFs in the GTI 2009 study compared with the 2019
study. The GTI 2009 study did not report leak and vented emissions separately for commercial meters. As the
GTI 2009 commercial meter EF is lower than the GTI 2019 EF, vented emissions may not have a noticeable
impact for commercial meters. Table 4 presents the commercial meter weighted average CH4 EF.
Table 4. Commercial Meter Weighted Average Population CH4 EF (kg/meter/yr)
Study
Data Points
Commercial Meter CH4 EF
GTI 2009
836
9.73
GTI 2019
337
57.4
Weighted Average EF
23
As noted previously, industrial meter EFs from the GTI 2009 study were not incorporated in the 2016 GHGI
customer meters updates, due to the limited sample size. EPA re-evaluated these data for the current
analyses. Industrial meter emissions from "Company B" in the GTI 2009 data account for 95% of the total
industrial meter emissions of the data set. The 2009 GTI study noted that the meter emissions measured for
Company B are largely due to pneumatic controller vented emissions. GTI 2009 also noted that the meter type
of Company B was similar to a metering and regulating station, versus a traditional meter (e.g., turbine,
rotary).
EPA discussed vented versus leak emissions with the GTI 2019 study authors, and the authors noted that for
the 2019 study data set (1) pneumatic controllers were not observed to be venting during the measurement
campaigns and (2) some regulators were observed to be venting during the measurement campaigns but their
emissions were variable and not quantified.
Table 5 presents the GTI 2009 industrial meters data, including the breakdown between leak and vented
emissions. Table 6 presents the resulting industrial meter weighted average CH4 EFs for the 2009 and 2019 GTI
studies, including separate weighted average EFs for leak and vented emissions. The vented emissions EF
presented in Table 6 incorporates the default assumption that vented emissions were zero during all GTI 2019
study measurements (i.e., it uses the data as-reported in each study). EPA is considering whether that is a
reasonable assumption, or if vented emissions data should be calculated only from the GTI 2009 study data or
with some other approach.
Page 4 of 10

-------
September 2020
Table 5. GTI 2009 Industrial Meters Data
Company
# Industrial Meters
Sampled
Leak CH4
Emissions (kg/yr)
Vented CH4
Emissions (kg/yr)
CH4 Emissions
(kg/yr)
A
7
411
0
411
B
7
734
170,341
171,075
C
0
0
0
0
D
2
29
6,616
6,646
E
22
609
9
618
F
8
735
0
735
Total
46
2,519
176,965
179,485
EF (kg/meter/yr)
55
3,847
3,902
Table 6. Industrial Meter Weighted Average Population CH4 EFs (kg/meter/yr)
Study
Data Points
Leak Emissions
ch4ef
Vented Emissions
CH4EF
Total CH4 EF
GTI 2009
46
55
3,847
3,902
GTI 2019
186
117.8
N/A
117.8
Weighted Average

105
763
868
4	Regional Variability and Time Series Considerations
The update under consideration for the 2021 GHGI does not include changes to the activity data. EPA is
considering applying the commercial and industrial meter EFs under consideration across the time series,
along with EIA activity data for commercial and industrial meters. This is consistent with the approach for
residential meters. If stakeholder information indicating that emission rates from commercial and industrial
meters have changed over the time series, and data were available to do so, EPA would consider using
different EFs overtime.
GTI 2019 recommends using region-specific EFs based on their findings of different meter set types in different
regions. The EIA dataset that EPA uses for industrial and commercial meter counts includes counts by state,5
and EPA could apply region-specific EFs to the meter counts from states within each region. Due to the limited
measurement data for each region, EPA is currently considering the use of national EFs and not regional EFs,
but seeks stakeholder feedback on this topic.
5	Preliminary National Emissions Estimates for Customer Meters in the
2021 GHGI
Based on the data sources and considerations discussed in Sections 3 and 4, this section summarizes the
approaches EPA is considering for the 2021 GHGI. As part of the update, EPA is proposing to no longer use a
single EF to apply to both commercial and industrial meters. EPA calculated preliminary national-level CH4
emission estimates for the update under consideration for commercial and industrial meters using multiple EF
scenarios to account for leak and vented emissions.
For commercial meters, because there is less data available to distinguish between vented and leak emissions
and the GTI 2019 study EF (which only measured leak emissions) is higher than the GTI 2009 study EF (which
measured leak and vented emissions), EPA evaluated two scenarios. Although the first scenario, which uses
5 http://www.eia.gov/dnav/ng/ng_cons_num_a_epg0_vn5_count_a.htm
Page 5 of 10

-------
September 2020
the GTI 2019 population EF, only accounts for leak emissions, EPA would not add commercial meter vented
emissions unless additional data from the GTI 2009 study can be ascertained or stakeholders provide
additional information. Commercial meter EF scenarios:
1.	GTI 2019 population EF in Table 1
2.	Weighted average EF in Table 4
For industrial meters, EPA evaluated two scenarios for leak emissions and two scenarios for vented emissions.
EPA would calculate total industrial meter emissions by summing leak plus vented emissions, dependent on
the scenario selected for each. Industrial meter EF scenarios:
1.	Leak - GTI 2019 population EF in Table 6
2.	Leak - Weighted average EF in Table 6
3.	Vented-GTI 2009 EF in Table 6
4.	Vented - Weighted average EF in Table 6
EIA provides activity data in the form of meter counts divided into industrial and commercial meters. In the
current GHGI, these values are summed. For the update under consideration, EPA used unique counts for
industrial and commercial meters and applied the respective EFs.
Table 7 summarizes the results for each scenario for the commercial and industrial meters update under
consideration for year 2018. Appendix A provides time series data for each scenario.
Note that the current 2018 values in the 2020 GHGI were based on an incorrect national industrial meter count
due to a spreadsheet error. Both the 2018 value reported in the 2020 GHGI and the corrected value are
included in the table below. The spreadsheet error only impacted year 2018 emissions.
Table 7. Year 2018 Customer Meters National Emissions Estimates Calculated by Various Approaches
Emissions Type
(Leak/ Vented)
EF Basis
EF
(Kg/meter/year)
AD (# meters)
2018 Emissions
(MT CH4)
Commercial Meters
Leak
GTI 2019
57.4
5,515,841
316,609
Leak + Vented
Weighted - GTI 2009 and
2019
23.43
5,515,841
129,227
Current GHGI - Leak
and Vented
GTI 2009 (commercial EF)
9.7
5,515,841
53,692
Industrial Meters
Leak
GTI 2019
117.8
184,943
21,786
Leak
Weighted - GTI 2009 and
2019
105
184,943
19,419
Vented
GTI 2009
3,847
184,943
711,489
Vented
Weighted - GTI 2009 and
2019
763
184,943
141,112
Current GHGI - Leak +
Vented
GTI 2009 (commercial EF)
9.7
251,484
2,448
Current GHGI - Leak +
Vented - CORRECTED
GTI 2009 (commercial EF)
9.7
184,943
1,800
Page 6 of 10

-------
September 2020
6 Requests for Stakeholder Feedback
EPA seeks stakeholder feedback on the approaches under consideration and the questions below.
1.	EPA seeks feedback on how to incorporate industrial meter venting emissions. The GTI 2019 study did
not measure venting emissions, but regulator venting emissions were observed (though variable in
nature). Table 6 presents weighted average EFs calculated from all study data, and which therefore has
a default assumption incorporated that venting emissions were zero during all GTI 2019 study
measurements. While regulator venting emissions were observed, the study does not have an
indication as to the magnitude of their impact.
2.	EPA seeks feedback on how to incorporate industrial meter leak emissions, including whether using
solely the GTI 2019 EF, or a weighted average EF (calculated from the combined dataset) is most
appropriate.
3.	EPA seeks feedback on how to incorporate leak and venting emissions for commercial meters,
including whether using the GTI 2019 EF or a weighted average EF (calculated from the combined
dataset) is most appropriate. EPA also seeks feedback on whether commercial meter vented emissions
should be supplemented with vented emissions data from industrial meters or if other data are
available to address vented emissions from commercial meters. Detailed leak and vented emission are
not available in the GTI 2009 study to determine the percent that each contributes. In addition, the GTI
2019 study EF (which only reflects leak emissions) is higher than the GTI 2009 EF (which includes leak
and vented emissions), which could suggest that vented emissions may not be a significant contributor
to commercial meter emissions.
4.	In addition to the specific leak and vented emissions questions above, EPA generally seeks feedback on
the most appropriate EFs to apply for commercial and industrial meters. This includes whether GTI
2019 study EFs should be applied, if weighted average EFs based on the GTI 2009 and 2019 studies are
more appropriate, if regional EFs should be considered, or if another approach or data source is
recommended.
5.	EPA seeks feedback on whether different EFs should be applied over the time series. EPA is considering
applying the same EFs, but could consider applying one EF to early years of the time series and a
different EF to recent years, with linear interpolation between if there is information available
indicating that the emission rate per meter has changed over the time series.
Page 7 of 10

-------
September 2020
Appendix A - Time Series Emissions and Activity Data for Various Approaches for Commercial and
Industrial Meters
Commercial and Industrial Meter CH4 Emissions by Various Approaches (MT cm/Year)
Emissions Type
(Leak/Vented)
EF Basis
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Commercial Meters
Leak
GTI 2019
243,162
250,106
253,117
256,286
260,246
266,135
270,941
273,305
289,554
287,585
287,621
286,796
290,696
295,735
295,033
Leak and Vented
Weighted-GTI
2009 and 2019
99,249
102,083
103,312
104,606
106,222
108,626
110,587
111,552
118,184
117,381
117,395
117,059
118,650
120,707
120,421
Current GHGI
Leak and Vented
GTI 2009
(commercial EF)
41,236
42,414
42,924
43,462
44,133
45,132
45,947
46,348
49,104
48,770
48,776
48,636
49,297
50,152
50,033
Industrial Meters
Leak
GTI 2019
25,721
25,507
24,693
24,699
23,906
24,667
24,273
27,666
26,645
26,897
25,946
25,566
24,257
24,210
24,627
Leak
Weighted- GTI
2009 and 2019
22,926
22,736
22,010
22,015
21,309
21,987
21,635
24,660
23,750
23,975
23,126
22,788
21,621
21,579
21,951
Vented
GTI 2009
839,973
833,003
806,408
806,600
780,725
805,569
792,685
903,504
870,173
878,406
847,325
834,915
792,170
790,627
804,261
Vented
Weighted-GTI
2009 and 2019
166,594
165,212
159,937
159,975
154,843
159,771
157,215
179,194
172,584
174,217
168,052
165,591
157,113
156,807
159,511
Current GHGI
Leak and Vented
GTI 2009
(commercial EF)
2,125
2,108
2,040
2,041
1,975
2,038
2,006
2,286
2,202
2,223
2,144
2,113
2,004
2,000
2,035
Page 8 of 10

-------
September 2020
Emissions Type
(Leak/Vented)
EF Basis
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
Commercial Meters
Leak
GTI 2019
298,367
302,692
304,724
312,505
305,502
304,310
305,357
307,457
308,383
310,738
313,038
314,248
315,620
316,609
Leak and Vented
Weighted-GTI
2009 and 2019
121,781
123,547
124,376
127,552
124,694
124,207
124,635
125,492
125,870
126,831
127,770
128,263
128,823
129,227
Current GHGI
Leak and Vented
GTI 2009
(commercial EF)
50,598
51,332
51,676
52,996
51,808
51,606
51,784
52,140
52,297
52,696
53,086
53,291
53,524
53,692
Industrial Meters
Leak
GTI 2019
24,293
22,833
23,358
26,510
24,458
22,704
22,300
22,308
22,652
22,634
22,186
22,245
21,787
21,786
Leak
Weighted- GTI
2009 and 2019
21,653
20,352
20,820
23,630
21,801
20,237
19,877
19,884
20,190
20,175
19,775
19,828
19,419
19,419
Vented
GTI 2009
793,355
745,678
762,832
865,760
798,744
741,446
728,254
728,528
739,746
739,172
724,542
726,466
711,504
711,489
Vented
Weighted-GTI
2009 and 2019
157,348
147,892
151,295
171,709
158,417
147,053
144,437
144,491
146,716
146,602
143,700
144,082
141,115
141,112
Current GHGI
Leak and Vented
GTI 2009
(commercial EF
2,007
1,887
1,930
2,191
2,021
1,876
1,843
1,843
1,872
1,870
1,833
1,838
1,800
1,800a
a - Corrected value for 2018 industrial meters emissions. The current 2018 values in the 2020 GHGI were based on an incorrect national industrial meter count due to a
spreadsheet error. The spreadsheet error only impacted year 2018 emissions. The 2018 value for CH4 from industrial meters presented in the 2020 GHGI was 2,448 kt CH4.
Please see section 5 of this memo for additional information.
Activity Data: Number of Commercial and Industrial Meters
Meter Type
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Commercial
4,236,280
4,357,252
4,409,699
4,464,906
4,533,905
4,636,500
4,720,227
5,064,384
5,152,177
5,139,949
4,236,280
4,357,252
4,409,699
4,464,906
4,533,905
Industrial
218,341
216,529
209,616
209,666
202,940
209,398
206,049
205,915
205,514
209,058
218,341
216,529
209,616
209,666
202,940
Meter Type
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
Commercial
5,198,028
5,273,379
5,308,785
5,444,335
5,322,332
5,301,576
5,319,817
5,356,397
5,372,522
5,413,546
5,453,627
5,474,701
5,498,603
5,515,841
Industrial
206,223
193,830
198,289
225,044
207,624
192,730
189,301
189,372
192,288
192,139
188,336
188,836
184,947
184,943a
a - Corrected value for 2018 industrial meters counts. The current 2018 values in the 2020 GHGI incorporated an incorrect national industrial meter count due to a spreadsheet
error. The spreadsheet error only impacted year 2018 data. The 2018 value for national number of industrial meters presented in the 2020 GHGI was 251,484. Please see
section 5 of this memo for additional information.
Page 9 of 10

-------
September 2020
Appendix B - Study Design Information
Meter Type
Measurement
Type
Number of Sources
Location and
Representativeness
EF Calculation
Method
GTI2009
Commercial
Hi Flow Sampler
measurements of
leaks and vents
836 meters at 6
companies
Spread across five
areas of the U.S.
Randomly selected
meters. The meters
tested equal
approximately
0.11% of the
meters in operation
at the 6 companies
GTI developed a
weighted average
EF based on
number of meters
tested.
Industrial
Hi Flow Sampler
measurements of
leaks and vents
46 meters at 5
companies
Spread across five
areas of the U.S.
Meters were
randomly selected.
GTI developed a
weighted average
EF based on
number of meters
tested
GTI2019
Commercial
Hi Flow Sampler
measurements of
leaks
337 meters at 10
companies
Spread across six
regions of the U.S.
Initial site for the
day was randomly
selected, efficient
route determined
from there.
GTI developed EFs
in multiple ways:
population, leaker-
only, regional
Industrial
Hi Flow Sampler
measurements of
leaks
186 meters at 10
companies
Spread across six
regions of the U.S.
Meters were
randomly selected
GTI developed EFs
in multiple ways:
population, leaker-
only, regional
Page 10 of 10

-------