of 9
Hydrocarbon Tracers Suggest Methane Emissions from Fossil
Sources Occur Predominately Before Gas Processing and That
Petroleum Plays Are a Signi
fi
cant Source
Ariana L. Tribby, Justin S. Bois, Stephen A. Montzka, Elliot L. Atlas, Isaac Vimont, Xin Lan,
Pieter P. Tans, James W. Elkins, Donald R. Blake, and Paul O. Wennberg
*
Cite This:
https://doi.org/10.1021/acs.est.2c00927
Read Online
ACCESS
Metrics & More
Article Recommendations
*
s
ı
Supporting Information
ABSTRACT:
We use global airborne observations of propane
(C
3
H
8
) and ethane (C
2
H
6
) from the Atmospheric Tomography
(ATom) and HIAPER Pole-to-Pole Observations (HIPPO), as
well as U.S.-based aircraft and tower observations by NOAA and
from the NCAR FRAPPE campaign as tracers for emissions from
oil and gas operations. To simulate global mole fraction
fi
elds for
these gases, we update the default emissions
con
fi
guration of C
3
H
8
used by the global chemical transport model, GEOS-Chem
v13.0.0, using a scaled C
2
H
6
spatial proxy. With the updated
emissions, simulations of both C
3
H
8
and C
2
H
6
using GEOS-Chem
are in reasonable agreement with ATom and HIPPO observations,
though the updated emission
fi
elds underestimate C
3
H
8
accumu-
lation in the arctic wintertime, pointing to additional sources of this gas in the high latitudes (e.g., Europe). Using a Bayesian
hierarchical model, we estimate global emissions of C
2
H
6
and C
3
H
8
from fossil fuel production in 2016
2018 to be 13.3
±
0.7 (95%
CI) and 14.7
±
0.8 (95% CI) Tg/year, respectively. We calculate bottom-up hydrocarbon emission ratios using basin composition
measurements weighted by gas production and
fi
nd their magnitude is higher than expected and is similar to ratios informed by our
revised alkane emissions. This suggests that emissions are dominated by pre-processing activities in oil-producing basins.
KEYWORDS:
ethane, propane, natural gas, methane, energy
INTRODUCTION
Many studies have diagnosed recent methane (CH
4
) trends
(both global and regional) using ethane (C
2
H
6
) atmospheric
ratio signatures. However, the rejection of C
2
H
6
by oil and gas
producers (in the U.S., and presumably in countries following
similar economic trends,
Figure S1
) results in an increase in
the mole fraction of C
2
H
6
in the natural gas pipelines. Thus, to
the extent that losses occur in the pipelines and at the end
users of natural gas, emissions of C
2
H
6
may not necessarily
directly re
fl
ect CH
4
emissions, adding additional uncertainty in
CH
4
emission estimates from natural gas operations. In
addition, a global uptick in hydraulic fracturing has shifted
production from dry to wet
fi
elds, resulting in an increase in
the ratios of both C
2
H
6
and C
3
H
8
to CH
4
,
1
further
complicating the use of the alkanes to diagnose the underlying
CH
4
emission sources.
2
4
Given the uncertainty in using C
2
H
6
alone as a tracer for
CH
4
emissions, we use both C
2
H
6
and propane (C
3
H
8
)to
diagnose whether signi
fi
cant CH
4
emissions from natural gas
and petroleum occur before gas processing. Unlike C
2
H
6
,C
3
H
8
has a much higher market value and therefore does not
undergo
rejection
.
5
Provided downstream losses are minimal
and the raw gas ratio of C
3
H
8
to CH
4
is known, C
3
H
8
can
provide a constraint for CH
4
emissions from raw, unprocessed
natural gas.
In this study, we employ global observations from aircraft,
including the 2009
2011 High-Performance Instrumented
Airborne Platform for Environmental Research (HIAPER)
Pole-to-Pole Observations (HIPPO)
6
and the 2016
2018
Atmospheric Tomography (ATom)
7
missions, which provide
vertical pro
fi
les of a variety of constituents, including C
2
H
6
and
C
3
H
8
, around the remote atmospheres of the globe. Together
with the large-scale chemical transport model GEOS-Chem,
we estimate global fossil emissions of C
2
H
6
and C
3
H
8
.
Received:
February 11, 2022
Revised:
May 18, 2022
Accepted:
May 18, 2022
Article
pubs.acs.org/est
© XXXX The Authors. Published by
American Chemical Society
A
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
Downloaded via CALIFORNIA INST OF TECHNOLOGY on June 14, 2022 at 16:48:29 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
MATERIALS AND METHODS
Observations from the National Oceanic and Atmos-
pheric Administration (N
OAA) Global Monitoring
Laboratory (GML).
Measurements of CH
4
,
8
10
C
2
H
6
, and
C
3
H
8
11
from
fl
ask air collected by the NOAA GML tower
12
and aircraft
13
near oil/natural gas basins were used as
unprocessed gas references. We refer to site locations using
state abbreviations. More information on data processing/
spatial coverage is included in
section S2, Table S1, and Figure
S5
. To better quantify geophysical variability and generate a
con
fi
dence interval in the correlation between C
2
H
6
and C
3
H
8
mole fractions, we implement a pairs bootstrap to generate
replicates of C
2
H
6
and C
3
H
8
observations. The CIs calculated
from the bootstrapped samples are much broader than those
calculated assuming the noise in the measurements is
dominated by analytical errors. This suggests that geophysical
noise induced by di
ff
erences in transport and chemistry
dominates the statistics. See
section S2.1
for more details.
FRAPPE Observations.
FRAPPE C130
fl
ight data were
taken within the Colorado Front Range between July 26 and
August 19, 2014. We accessed the data on October 6, 2021
from www.air.larc.nasa.gov from the WAS C130 merge
fi
le.
Our data processing for FRAPPE is similar to our methods for
the NOAA in situ samples. A spatial illustration of FRAPPE
observations is shown in
Figure S6
.
HIAPER Pole-to-Pole Observations and ATmospheric
Tomography Data.
The HIPPO campaign was a sequence of
fi
ve global measurement campaigns which sampled from near
the North Pole to the coastal waters of Antarctica, covering
di
ff
erent seasons between 2009 and 2011. Similarly, ATom
took place from 2016 and 2018. Flight paths of HIPPO and
ATom campaigns are illustrated in
Figure S22
, and speci
fi
c
details about the data sources are included in
section 3
of the
Supporting Information (SI).
Only data observed at >20
°
north were used since the
majority of emissions of these short-lived gases of interest lie in
the northern hemisphere. The lifetime of C
3
H
8
and C
2
H
6
are
on the order of a few months or shorter during the summer,
and the time it takes for mixing between the northern to
southern hemispheres is on the order of a year,
14
so the mole
fraction of these gases of interest is very low in the southern
hemisphere.
Because C
2
H
6
and C
3
H
8
are relatively short-lived gases, their
abundance in the stratosphere is low and poorly connected to
the underlying
fl
uxes. To exclude stratospheric observations,
we use N
2
O (Panther/UCATS instrument), which is inert and
generally well-mixed in the troposphere but is destroyed in the
stratosphere by photolysis and reaction with O
1
D.
15
Thus, we
exclude from our analysis data with low N
2
O mole fraction
(
Figure S23
).
As our focus in this analysis is quantifying the global
emissions of these gases, we exclude from our analysis data
where local
fl
uxes substantially in
fl
uence the mole fraction of
these alkanes. We use a simple land and altitude constraint and
HCN as a tracer to remove plumes from highly local sources
(including both energy infrastructure and wild
fi
res,
Figures
S24
S26
). We also exclude regions and times where the
lifetime of the alkanes is very short and thus regional/local
sources dominate the variance. Thus, we do not analyze the
aircraft summer data (results for the summer are shown in the
SI
) or data in the subtropics, where the alkane distribution is
very sensitive to transport from the extratropics where most
emissions of C
2
H
6
and C
3
H
8
originate. To exclude subtropical
air, we only analyze measurements with tropopause pressure
above 100 hPa (about 5% of the data were excluded under this
constraint) for both ATom and HIPPO, which was su
ffi
cient
to reduce the in
fl
uence of tropical intrusions. See
Table S2
for
a comprehensive outline of the
fi
lters we use.
As in other studies,
16
,
17
we use potential temperature (
θ
,in
units of Kelvin) in our analysis as a zonal coordinate. Potential
temperature is conserved following adiabatic
fl
ow, and in the
extratropics, variability within large-scale circulation can be
well captured using this coordinate system. As a result, trace
gases that have long lifetimes compared to synoptic-scale
meteorology, which has a horizontal length scale on an order
of 1000 km or more and a time scale of about 10 days,
14
,
18
will
be well correlated with
θ
. Using
θ
as a dynamical coordinate
allows us to more accurately compare low spatial resolution
GEOS-Chem simulations with the aircraft in situ measure-
ments (compared with simply using altitude and latitude
coordinates,
Figure S27
). Potential temperature is not well-
correlated with trace gases in the tropics or boundary layer,
where moist convection and surface-drag-driven turbulence
can result in nonunique pairs or when the photochemical
lifetimes are short (summer).
GEOS-Chem Simulations.
We simulated HIPPO and
ATom measurements using the GEOS-Chem
classic
global
3-D chemical transport model with default settings (details
about the simulations and emissions are provided in the SI,
section 4
). We use the same constraints as the aircraft
observations, except we use a boundary layer height parameter.
As described below, we use a Bayesian model to provide a best
estimate for global emissions of C
2
H
6
and C
3
H
8
and their
uncertainty. One contribution to the error estimate is transport
errors in GEOS-Chem. To capture some of the uncertainty in
the transport
fi
eld, we sample the GEOS-Chem model several
days before and after the in situ sampling time along the
aircraft
fl
ight path, which we refer to as
synoptic replicates
.
Finally, all GEOS-Chem simulations of C
3
H
8
and C
2
H
6
were
interpolated on the vertical coordinate using
θ
to the aircraft
measurements. As expected, GEOS-Chem synoptic replicates
show less consistency in latitude (
Figure S27
), providing
support for using
θ
as an analysis coordinate.
Bayesian Inference.
We wish to quantify the global
emissions of C
3
H
8
and C
2
H
6
using the observed mole fraction
of these alkanes during ATom and HIPPO. The ambient mole
fraction of C
3
H
8
and C
2
H
6
is most sensitive to their total
northern hemisphere emissions during the winter/fall/spring
when there is decreased sunlight/oxidation. As such, we
assume di
ff
erences between the GEOS-Chem simulations and
the aircraft observations can be largely attributed to the
underlying emissions grid, such that
α
a
gcs
(1)
where
a
is the aircraft C
2
H
6
or C
3
H
8
mole fraction, gcs is the
GEOS-Chem simulation of C
2
H
6
or C
3
H
8
mole fraction, and
α
is a scalar that represents the missing emissions of C
3
H
8
and
C
2
H
6
from default emissions. We developed a Bayesian
hierarchical model to estimate the missing emissions, where
eq 1
forms the basis of our model. Our model only uses the
GEOS-Chem simulated alkane mole fraction data (synoptic
replicates), the alkane mole fractions observed by the aircraft,
tropopause height, and UTC time. Our complete statistical
model and its development, priors (
Figures S37
S41
), as well
as the software used, are included in the SI,
section 5
.
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
B
Hydrocarbon Percent Composition Literature Com-
pilation and Bootstrapping.
We gather literature measure-
ments of hydrocarbon composition from unprocessed gas from
oil- and gas-producing basins in the U.S. and around the globe
to calculate emission ratios. Summary statistics of the percent
composition by region and the corresponding literature source
is included in
Tables S4 and S5
(SI). Gas composition varies
signi
fi
cantly across basins, so we perform bootstrap calcu-
lations for data samples within each basin separately. For each
basin, we draw random pairs of hydrocarbon composition
measurements (CH
4
,C
2
H
6
, and C
3
H
8
) for the size of the data
set and then take the mean and repeat this 10,000 times. We
use these bootstrap samples in subsequent calculations (
eq 2
)
to arrive at emission ratios (
Figure 6
).
RESULTS AND DISCUSSION
CH
4
Leaks from U.S. Energy Activities Are Dominated
by Emissions from Unprocessed Natural Gas.
C
3
H
8
and
C
2
H
6
are highly correlated at the NOAA sites (
Figure 1
). The
data shown were obtained across 2
11 years and include
tower and aircraft data (
Table S1
). The cross plot of C
3
H
8
and
C
2
H
6
illustrates two distinct chemical regimes, similar to those
described by Parrish et al.
19
Above 1 ppb C
3
H
8
,the
distribution is nearly linear, consistent with the mixing of
fresh non-photochemically aged emissions into the background
atmosphere. At mole fractions below 1 ppb, a second regime is
de
fi
ned by mixing of the aged emissions (the lifetime of C
3
H
8
is much less than that of C
2
H
6
). To explore the characteristics
of unprocessed natural gas emissions, we study the ratio of
these gases within the 50th highest percentile of C
3
H
8
for the
combined sites. Varying this demarcation
±
10% negligibly
a
ff
ects the linear
fi
t(
Figure S10
). We
fi
nd that the ratio of
C
3
H
8
to C
2
H
6
in the linear regime to be [0.63, 0.70] (ppb/
ppb, 95% CI), and FRAPPE observations show a similar trend
at [0.76, 0.87] (
Figure S9
).
Within the fresh emission chemical regime, the NOAA
C
3
H
8
/C
2
H
6
ratio changes minimally before and after 2012
([0.62, 0.67] and [0.63, 0.71], respectively, ppb/ppb 95% CI,
Figure S9
) and over the entire time series (
Figure S19
). An
unchanging C
3
H
8
/C
2
H
6
ratio over time across the U.S. despite
large changes in the C
3
H
8
/C
2
H
6
ratio in processed gas during
the same years (
Figure S1
) suggests that the majority of the
alkane (and likely CH
4
) emissions occur before the gas
processing stage. (During gas processing, most of the C
3
H
8
and sometimes much of the C
2
H
6
are separated from the raw
gas.) Conversely, post-processing (pipeline) composition of
C
2
H
6
at Playa del Rey in Southern California follows rejection
trends, where the C
3
H
8
/C
2
H
6
ratio has decreased by 8% from
2008 to 2018 and in recent years is about 18% lower in
magnitude compared to the NOAA ratio (
Figure S4
).
(Processed gas in California is a good representation of typical
gas composition of domestic and globally imported consumer-
grade gas.
20
) Our results are in agreement with Rutherford et
al.
s U.S.-based model for CH
4
emissions, which
fi
nds that
unintentional emissions from the production segment (namely,
liquid storage tanks and other equipment leaks) are the largest
contributors to divergence with the EPA
s GHGI.
21
We use the same chemical aging approach to construct a
background for NOAA and FRAPPE CH
4
observations
(
Figures S7 and S11
). Since we only focus on the linear part
of the chemical aging distribution, our analysis is not terribly
sensitive to how the CH
4
anomaly is determined (it simply
produces varying intercepts,
Figure S10
). Consistent with the
analysis of Lan et al.
2
(see
Figures S20 and S21
), the ratio of
C
3
H
8
and C
2
H
6
to CH
4
has increased with time, re
fl
ecting a
growing importance of oil exploration on CH
4
emissions in the
U.S.
C
3
H
8
is demonstrated to be a useful tracer that constrains
oil- and gas-related CH
4
emissions. Given that the
fi
tofC
3
H
8
versus CH
4
(and C
2
H
6
versus CH
4
) have similar precision over
the whole record, using C
3
H
8
as a tracer likely separated
nearby competing non-oil and gas CH
4
emissions. However, if
there were to be non-oil and gas CH
4
emissions that were
spatially coherent to the NOAA observation sites, our C
3
H
8
versus CH
4
and C
2
H
6
versus CH
4
emission ratios would be
impacted. Investigating and potentially separating spatially
coherent emissions of nonfossil origins would be the topic of
future studies.
NOAA observations at Oklahoma ARM site are especially
impacted by nearby unprocessed gas emissions, as C
2
H
6
and
C
3
H
8
correlations with CH
4
have less noise compared to other
sites (
Figures S12 and S13
). While the ratios of C
3
H
8
and
C
2
H
6
versus CH
4
have increased by 50% at NOAA Oklahoma
site since 2010, we
fi
nd that both C
2
H
6
and C
3
H
8
versus CH
4
ratios are fractionally increasing at the same rate (
Figure 2
).
That the atmospheric C
2
H
6
and C
3
H
8
increase fractionally the
same suggests that the ratio of the alkanes in the reservoirs
producing these emissions do not change signi
fi
cantly over the
time of this record. Below, we use the 2017 average C
3
H
8
/CH
4
and C
2
H
6
/CH
4
from NOAA Oklahoma site ([0.060, 0.061]
and [0.086, 0.088] ppb/ppb, respectively,
Figure S21
)to
compare to ratios between our emission estimates for C
3
H
8
and C
2
H
6
to published estimates of CH
4
emissions from oil
and gas exploration.
Default GEOS-Chem Simulations Underestimate C
3
H
8
Compared to Aircraft Observations.
We compare the
cross plot of C
3
H
8
to C
2
H
6
from the HIPPO and ATom
aircraft measurements and GEOS-Chem simulations to the
NOAA measurements (
Figure 3
). As expected, both the
aircraft observations and GEOS-Chem simulations fall under
Figure 1.
Measurements of C
2
H
6
and C
3
H
8
from ongoing NOAA
GML tower and aircraft sites (
Table S1
) from 2005 to 2018. The data
follow the photochemical aging distribution described in Parrish et al.,
where the data below 1 ppb C
3
H
8
are a
ff
ected by photochemically
aged emissions and mixing processes. As such, we only study the ratio
of these gases in the 50th highest percentile (everything above 1 ppb
C
3
H
8
) that would indicate fresh emissions. After this
fi
ltering, two
sites, Northwestern CO and Western UT (site codes NWR and
UTA), did not have any data in the fresh emission regime and are not
included in further analysis (more detail in
Figure S8
).
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
C
the photochemically aged emissions part of the NOAA
distribution. While the aircraft data overlay the NOAA
measurements almost perfectly (especially in the winter
when the lifetimes of both gases are the longest), GEOS-
Chem underestimates C
3
H
8
, particularly over the Atlantic
curtain (
Figures 3
and
4
). The same conclusion is drawn for
HIPPO time periods (
Figures S31, S32, and S34
). Because the
atmospheric lifetimes of C
3
H
8
and C
2
H
6
are di
ff
erent and vary
seasonally due to the much higher concentrations of OH in the
summer, our estimate of global C
2
H
6
and C
3
H
8
emissions from
GEOS-Chem comparisons are sensitive to the a priori spatial
distribution of these emissions.
Relative to C
2
H
6
, the default GEOS-Chem v13.0.0 C
3
H
8
emissions result in a much larger underestimate of C
3
H
8
mole
fractions over the Atlantic transect compared to the Paci
fi
c
(
Figures 3
and
4
), implying an underestimate over North
America. This pattern is most clearly visible in the summer
when the C
3
H
8
lifetime is short. In contrast, the default C
2
H
6
emissions produce a good simulation of the ATom data
(within 5%) over both ocean basins (
Figures 3
and
4
). As such,
we use the mean ratio observed in the linear regime of the
NOAA data (0.67 C
3
H
8
/C
2
H
6
, ppb/ppb
Figure S9
) as the
default global ratio of their emissions to update GEOS-Chem.
In mass units (as used in the GEOS-Chem emissions), this is
0.99 kg C
3
H
8
/kg C
2
H
6
. Given the remarkable coherence in
both the large-scale
fi
elds from the aircraft over both the
Atlantic and Paci
fi
c transects and in the NOAA data, the
spatial distribution of the emissions ratios for both gases must
be very similar upwind of the Paci
fi
c (e.g., Asia) and the
Figure 2.
Yearly correlation between NOAA hydrocarbon versus CH
4
anomaly in Oklahoma. We show the percent change of anomalies/
year with respect to the mean hydrocarbon and methane anomalies.
The trend for C
3
H
8
/CH
4
is 7.13
±
1.44% with an
R
2
of 0.71. The
trend for C
2
H
6
/CH
4
is 5.87
±
1.26% with an
R
2
= 0.69. The
variability in the trend comes from the standard error of a linear
regression. The variability in the individual points comes from the
95% con
fi
dence interval of a pairs bootstrap of the alkanes and CH
4
anomalies. (We ran a pairs bootstrap for co-measurements of C
3
H
8
and
Δ
CH
4
and compute the slope of the correlation for each
bootstrap sample and repeated this for every year in the data; please
see the
Materials and Methods
section.) This trend in units of ppt/
ppb/year is shown in
Figure S21
.
Figure 3.
Comparison of C
3
H
8
versus C
2
H
6
for NOAA, ATom aircraft, and GEOS-Chem simulations during fall/winter seasons. NOAA
photochemically aged measurements (all sites, 2005
2018), as explained in the text, are shown on the heat map (colored by the number density of
data). The spring/summer seasons are included in
Figures S29, S30, and S33
. HIPPO is shown in
Figure S34
.
Figure 4.
Impact of revised C
3
H
8
emissions on GEOS-Chem
simulation. Combined Paci
fi
c and Atlantic transects for ATom 4
aircraft campaign, which took place during Spring 2018, are shown in
gold. The GEOS-Chem simulation using default C
3
H
8
emissions are
shown in blue and orange, referring to the Paci
fi
c and Atlantic
transects, respectively. The GEOS-Chem simulation after implement-
ing the revised C
3
H
8
emissions is shown in green. The rest of the
ATom and HIPPO campaigns are shown in
Figures S35 and S36
.
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
D
Atlantic (North America). As such, in our revised emission
fi
elds for C
3
H
8
, we simply used the default C
2
H
6
emissions
con
fi
guration used by GEOS-Chem v13.0.0. This scaling
substantially altered the spatial distribution of C
3
H
8
emissions
(
Figure S28
). The e
ff
ect on the C
3
H
8
simulation is shown in
Figure 4
using ATom 4 as an example, where updating the
emissions resulted in a much better agreement between
GEOS-Chem C
3
H
8
and aircraft measurements (other
campaigns are included in
Figures S35 and S36
). Although
simulations are greatly improved using the revised emissions, it
appears there is a missing high latitude source of C
3
H
8
and
C
2
H
6
(
Figures S47, S48, S54, and S55
).
Bayesian Model Suggests Decadal Increase in Global
C
2
H
6
and C
3
H
8
Anthropogenic Fossil Emissions.
The
results of our Bayesian inference were satisfactory. We had
good sampling of our posterior, and the sampling diagnostics
were excellent (
Figures S43, S44, S50, and S51
). We
performed tests that veri
fi
ed our inference procedure could
capture the ground truth (using simulated data for which the
ground truth is known), and that our posterior was more
concentrated around the ground truth than the prior (
Figure
S42
). Furthermore, our model could generate the measured
data reasonably well; the majority of the measured aircraft data
fell into the 30th and 50th percentile of the simulated Bayesian
model data (
Figures S47, S48, S54, and S55
). The exception to
this was the summer season, where the Bayesian model does
not capture the measured aircraft data well. This is expected,
since during the summer we do not observe a robust
relationship between potential temperature and C
3
H
8
or
C
2
H
6
. The model remained robust even when varying the
sensitivity to low mole fractions of alkanes with tropical origin
(
Figures S58
S60
).
During ATom 2 (winter) Atlantic curtain, the GEOS-Chem
simulations poorly capture the observed C
2
H
6
and C
3
H
8
at low
potential temperature compared to the aircraft. These
measurements are samples obtained at low altitude, high
latitude, and cold temperatures (
Figure S49
). During the
winter, these arctic airmasses are often characterized by
stagnant conditions, with less mixing with the midlatitudes.
22
As a result, emissions that occur at high latitudes during the
winter can be trapped there unless the zonal
fl
ow is disrupted.
Additionally, emissions of C
2
H
6
and C
3
H
8
near the arctic
during the winter will oxidize more slowly relative to
midlatitudes due to the cold temperature and minimal sunlight.
These conditions result in high C
2
H
6
and C
3
H
8
mole fraction
over the arctic relative to remote midlatitude chemical regimes.
The measurements subject to arctic conditions during ATom
were too few to make a substantial impact on the overall
Bayesian emission estimates, but during HIPPO winter
fl
ights,
few samples were obtained, resulting in a large bias toward
arctic data. We place much more weight on the ATom winter
C
3
H
8
analysis estimate of global C
3
H
8
emissions since the
arctic represents only a small fraction of the atmosphere.
It is likely that GEOS-Chem v13.0.0 is missing a high
latitude emissions source (
Figures S47
S49, S54, and S55
).
Underestimation of C
3
H
8
and C
2
H
6
at high latitudes is
consistent with other studies, which found that fossil fuel
emissions from Eurasia accounted for the largest under-
estimation.
23
It is possible that emissions from northern
Europe may account for this discrepancy, as fossil emissions
were found to be underestimated,
23
and our revised C
3
H
8
emissions decreased in that region after implementing the
emission C
2
H
6
proxy (
Figure S28
). This, combined with a
relatively lower number of HIPPO aircraft observations at
lower latitudes, results in a substantial positive bias on the
overall Bayesian emissions estimate for C
3
H
8
during winter
2009 (
Figure S62
).
We report our Bayesian est
imates for each seasonal
campaign and ocean transect during 2009
2011 and 2016
2018 and what the GEOS-Chem v13.0.0 default emissions
grids should be scaled by, according to our analysis (
Figures
S45, S46, S52, and S53
and
SI section 5.8
). We estimate global
emissions of C
2
H
6
and C
3
H
8
from fossil fuel production from
2016 to 2018 to be 13.3
±
0.7 (95% CI) and 14.7
±
0.8 (95%
CI) Tg/year, respectively. Our results compare well to other
studies (
Figure 5
and
Figure S62
). Our estimates suggest
emissions of C
2
H
6
have increased by about 15% from 2010 to
2017 when comparing the mean revised C
2
H
6
emissions
during those time periods (
Figure S61
). Emissions of C
3
H
8
are
calculated to have increased more (65%,
Figure S62
), but this
estimate is highly uncertain due to the few samples obtained
during HIPPO and the impact of the Arctic winter pooling in
both campaigns. Nevertheless, these increases are consistent
with greater oil production emissions contribution. Similarly,
Helmig et al., which used data from a global surface network
and atmospheric column observations, found about a 22%
increase in C
2
H
6
emissions between 2009 and 2014.
24
Oil Exploration Plays a More Signi
fi
cant Role in
Global CH
4
Compared to Dry Gas.
We calculate an
emission ratio of C
3
H
8
/CH
4
for
n
basins or countries using the
following equation:
i
k
j
j
j
j
j
y
{
z
z
z
z
z
i
k
j
j
j
j
j
j
j
i
k
j
j
j
j
j
j
y
{
z
z
z
z
z
z
i
k
j
j
j
j
j
y
{
z
z
z
z
z
y
{
z
z
z
z
z
z
z
=
×
×
=
E
C
C
P
P
C
C
i
n
i
C
C
i
i
3
1
1
DNG
tot
DNG
tot
3
1
i
i
1
1
(2)
Figure 5.
Global revised ethane anthropogenic fossil emissions
compared to other studies. Our emissions estimate in 2016
2018
(during ATom) and 2009
2011 (during HIPPO) includes our
revised emissions for winter, fall, and spring seasons that we
determined with our Bayesian model during each season. As discussed
in the text, fewer samples were obtained during HIPPO, resulting in a
sampling bias that we test by restricting observations and simulations
to
±
300 K potential temperature (
Figures S56 and S57
). This test
a
ff
ects the estimate about
±
1 Tg during 2010
2011 but a
ff
ects our
estimate by up to 12 Tg in 2009. We compare our revised emissions
to the default emissions from GEOS-Chem v13.0.0. The studies
included here
23
,
25
27
represent anthropogenic fossil emissions, except
for the work by Dalsøren et al., which also includes biofuel,
agriculture, and waste. We obtained the CEDS CMIP6 estimate
from Dalsøren et al. Our emissions estimates do not include biomass
burning or biofuels. Propane emissions are included in
Figure S62
.
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
E
where
P
DNG
is the dry natural gas production (in million
tonnes),
C
1
and
C
3
are the bootstrapped samples of measured
hydrocarbon fractions in raw natural gas samples (in mass%,
details on the bootstrapping in the
Materials and Methods
section), and tot is the sum of the bootstrapped samples of
measured hydrocarbon fractions for CH
4
,C
2
H
6
, and C
3
H
8
.
(Note that our
()
E
C
C
3
1
emission ratios are inherently weighted
by natural gas production by basin,
P
DNG
.)
()
E
C
C
2
1
is calculated
similarly. We refer to the emission ratio in
eq 2
as the
literature
emission ratio, since we combine a variety of
natural gas composition measurements for wet and dry basins.
We calculate a global literature emission ratio using hydro-
carbon and dry natural gas production data from the top 5
producing natural gas basins around the world that made up
50% of the total natural gas production in 2019.
28
,
29
We also
calculate a U.S. literature emission ratio using the top 7
natural-gas-producing basins that account for 86% of total U.S.
natural gas production.
30
,
31
We show the relative production of
the top global and U.S. basins used in our analysis in
Figure S3
.
Additional summary statistics and sources for the composition
measurements are included in
Tables S4 and S5
.
Separately, we calculate an
observationally informed
emission ratio (OIER)
by taking the ratio between our
revised C
3
H
8
emissions with literature estimates of CH
4
emissions from oil and natural gas processes. The OIER for
C
2
H
6
/CH
4
is calculated similarly. Previous studies have
constrained global CH
4
emissions from natural gas/petroleum
systems to range from 63 to 91 Tg/year.
25
,
32
37
Given our
estimate for revised C
2
H
6
and C
3
H
8
emissions, this implies a
mean alkane ratio of 100:[8.0,10.0]:[6.5, 7.3] molar % (CH
4
/
C
2
H
6
/C
3
H
8
) in 2016
2018. We compare our literature ratio
with several OIER using global estimates in
Figure 6
and with
U.S. estimates in
Figure S63
. The small abundance of spatial
and temporal literature measurements of raw gas composition
throughout basins most a
ff
ect the
fi
nal uncertainty in the
emission ratio comparison.
Unprocessed dry gas has a smaller C
3
H
8
/CH
4
and C
2
H
6
/
CH
4
ratio compared to unprocessed associated gas from oil-
producing basins (
wet
gas). A greater contribution of
emissions from dry basins would decrease the magnitude of
the overall literature ratio, assuming minimal C
3
H
8
leakage
after separation from raw gas, given the high market value of
C
3
H
8
. In the U.S., the basin with the highest gas production is
the Appalachian (East Coast) (
Figure S3
). If CH
4
leaks were
proportional to production, we would expect a
dry
(small)
emission ratio that resembl
es the composition of the
Appalachian region (6% mass/mass of C
3
H
8
/CH
4
, calculated
from
Table S5
). However, we
fi
nd the production-weighted
literature ratio to be much larger than expected (15% mass/
mass for C
3
H
8
/CH
4
,
Figure S63
). The second-largest gas-
producing region, the Permian (Southwestern U.S.), is also the
largest oil producer in the U.S. and vastly overpowers the
Appalachian in terms of oil production (
Figure S3
). The
magnitude of our production-weighted raw gas
literature
emission ratios suggest a signi
fi
cant contribution from wet gas
and emissions that are biased toward oil-producing basins. We
fi
nd similar results for global emission ratios (
Figure 6
).
The Global Carbon Project CH
4
emissions estimate implies
an OIER that is dry relative to the production-weighted raw
gas ratio (literature ratio) (
Figure 6
). Instead, the IEA (76 Tg/
year) and Scarpelli et al. (66 Tg/year) CH
4
emission estimates
yield an OIER that is within a few percent of the production-
weighted raw gas literature ratio, given our revised C
3
H
8
and
C
2
H
6
emissions. Both of those studies estimate oil production
emissions to have a relatively higher contribution to the global
footprint compared to dry gas production.
The FRAPPE and NOAA Oklahoma observed emission
ratios compare well to the global OIER (
Figure 6
) and U.S.
OIER (
Figure S63
), suggesting high emissions from oil
production. Indeed, there are substantial oil production
activities (
Figures S15
S18
) surrounding the NOAA Oklaho-
ma and FRAPPE observation sites. Increasing trends in the
Oklahoma emission ratios are consistent with production
trends: oil production in Oklahoma tripled from 2010 to
2017
38
,
39
(compared to doubling of gas production), and in
2020, Oklahoma was the fourth-largest oil producer in the
U.S.
40
(Note that Oklahoma was not included in calculations
for the literature ratio, since Oklahoma natural gas production
does not rank in the top 7.) Several factors may reduce
incentive or the ability for oil producers to capture associated
Figure 6.
Global literature and observationally informed emission
ratios (OIER) C
3
H
8
/CH
4
and C
2
H
6
/CH
4
. The
weighted raw gas
ratio
in the
fi
gure represents the
literature ratio
described in the
text, calculated using
eq 2
. OIER, ratios between our revised C
2
H
6
and C
3
H
8
emissions and literature CH
4
emission estimates, are shown
for several literature CH
4
estimates, including IEA (76.4 Tg/year),
34
Scarpelli et al. (65.7 Tg/year),
33
and Global Carbon Project bottom-
up estimate (128 Tg/year, 2008
2017 average).
32
The variability in
the literature ratio is attributed to the 95% CI of pairs bootstrap
samples of hydrocarbon composition measurements (see text for
more detail). The variability in the OIER is attributed to the 95% CI
of our revised C
3
H
8
and C
2
H
6
emission estimates. We also compare
C
3
H
8
/CH
4
and C
2
H
6
/CH
4
correlations from in situ observations,
including NOAA observations from Northern Oklahoma (2017
average from
Figure S21
, units of kg/kg) and FRAPPE observations
from Northern Colorado (2014 from
Figure S9
, units of kg/kg). The
variability in the NOAA ratio is relatively low because it is calculated
from a multiyear average slope, and the error in the slope is low (see
Figure S21
, left). The variability in the FRAPPE ratio is relatively high
because we use the 95% CI derived directly from our bootstrap
samples, as described in the
Materials and Methods
section.
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
F
natural gas byproducts, including low market prices and
lagging pipeline infrastructure.
41
Since our
fi
ndings suggest that CH
4
losses are likely greater
and biased toward oil-producing sites, a signi
fi
cant fraction of
bottom-up estimates of CH
4
emissions may be misallocated to
dry CH
4
production when they should instead be included
with the oil production sector. Correctly attributing CH
4
emissions to oil production would increase the greenhouse
gas footprint of petroleum-based transportation, while
decreasing the greenhouse gas emissions ascribed to natural-
gas-powered power plants. At a minimum, the CO
2
equivalent
footprint of the global transportation sector would increase by
roughly 5%, using IEA
s estimate of 76 Tg/year CH
4
emissions
from oil and natural gas and recent transportation CO
2
emission estimates (
section 6.2
, SI).
36
,
42
This estimate will
only increase when accounting for vented and
fl
ared losses of
associated natural gas that is not accounted for in marketed
associated gas (which we use to calculate these numbers).
ASSOCIATED CONTENT
*
s
ı
Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.est.2c00927
.
U.S. and global economic and production trends of oil
and natural gas; statistical methods and chemical aging
approach to NOAA and FRAPPE observations; methane
anomaly determination for NOAA and FRAPPE
observations; yearly methane anomaly ratios for
NOAA observations; ATom and HIPPO aircraft
campaign
fi
ltering methods; GEOS-Chem simulation
results after revising propane emissions; Bayesian
inference results; emission ratio calculation details;
analysis of reallocating methane emissions to the
transportation sector (
PDF
)
AUTHOR INFORMATION
Corresponding Author
Paul O. Wennberg
Division of Engineering and Applied
Science and Division of Geological and Planetary Sciences,
California Institute of Technology, Pasadena, California
91125, United States;
orcid.org/0000-0002-6126-3854
;
Email:
wennberg@caltech.edu
Authors
Ariana L. Tribby
Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena,
California 91125, United States;
orcid.org/0000-0002-
6435-4575
Justin S. Bois
Division of Biology and Biological Engineering,
California Institute of Technology, Pasadena, California
91125, United States
Stephen A. Montzka
National Oceanic and Atmospheric
Administration, Global Monitoring Laboratory, Boulder,
Colorado 80305, United States
Elliot L. Atlas
Rosenstiel School of Marine and Atmospheric
Science, University of Miami, Miami, Florida 33149, United
States;
orcid.org/0000-0003-3847-5346
Isaac Vimont
National Oceanic and Atmospheric
Administration, Global Monitoring Laboratory, Boulder,
Colorado 80305, United States; Cooperative Institute for
Research in Environmental Sciences, University of Colorado
Boulder, Boulder, Colorado 80309, United States
Xin Lan
National Oceanic and Atmospheric Administration,
Global Monitoring Laboratory, Boulder, Colorado 80305,
United States; Cooperative Institute for Research in
Environmental Sciences, University of Colorado Boulder,
Boulder, Colorado 80309, United States
Pieter P. Tans
National Oceanic and Atmospheric
Administration, Global Monitoring Laboratory, Boulder,
Colorado 80305, United States
James W. Elkins
National Oceanic and Atmospheric
Administration, Global Monitoring Laboratory, Boulder,
Colorado 80305, United States; Cooperative Institute for
Research in Environmental Sciences, University of Colorado
Boulder, Boulder, Colorado 80309, United States;
orcid.org/0000-0003-4701-3100
Donald R. Blake
Department of Chemistry, University of
California
Irvine, Irvine, California 92697, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.est.2c00927
Notes
The authors declare no competing
fi
nancial interest.
ACKNOWLEDGMENTS
This work was supported by the Resnick Sustainability
Institute, including computations conducted in the Resnick
High Performance Computing Center. A.L.T. received funding
from NSF Award No. DGE-1745301. NOAA support was
provided for HIPPO by NSF Award No. AGS-0628452;
California Institute of Technology support for ATom was
provided by NASA Grant Award No. NNX15AG61A. NOAA
support for ATom was provided by NASA EVS2 Award No.
NNH17AE26I; additional support was provided by NASA
Upper Atmosphere Research Program award No. NNH13A-
V69I. NOAA laboratory and salary support is from NOAA
Climate Change Program. S.A.M. acknowledges funding in
part from NOAA Climate Program O
ffi
ce
s AC4 Program.
NOAA
fl
ask sampling and technical support was provided by
Dr. Fred Moore of NOAA/CIRES. Additional technical
support was provided by C. Siso, B. Miller, M. Crotwell, C.
Sweeney, A. Andrews, J. Higgs, D. Ne
ff
,J.Ko
fl
er, K. McKain,
M. Madronich, P. Handley, and S. Wolter. We thank IHS
Markit for providing PointLogic economic data.
REFERENCES
(1) U.S. Energy Information Administration.
Where our natural gas
comes from
;
https://www.eia.gov/energyexplained/natural-gas/where-
our-natural-gas-comes-from.php
(accessed 2022-03-18).
(2) Lan, X.; Tans, P.; Sweeney, C.; Andrews, A.; Dlugokencky, E.;
Schwietzke, S.; Kofler, J.; McKain, K.; Thoning, K.; Crotwell, M.;
Montzka, S.; Miller, B. R.; Biraud, S. C. Long-Term Measurements
Show Little Evidence for Large Increases in Total U.S. Methane
Emissions Over the Past Decade.
Geophys. Res. Lett.
2019
,
46
(9),
4991
4999.
(3) Turner, A. J.; Frankenberg, C.; Wennberg, P. O.; Jacob, D. J.
Ambiguity in the Causes for Decadal Trends in Atmospheric Methane
and Hydroxyl.
Proc. Natl. Acad. Sci. U. S. A.
2017
,
114
(21), 5367
5372.
(4) Kort, E. A.; Smith, M. L.; Murray, L. T.; Gvakharia, A.; Brandt,
A. R.; Peischl, J.; Ryerson, T. B.; Sweeney, C.; Travis, K. Fugitive
Emissions from the Bakken Shale Illustrate Role of Shale Production
in Global Ethane Shift: Ethane Emissions From the Bakken Shale.
Geophys. Res. Lett.
2016
,
43
(9), 4617
4623.
(5) U.S. Energy Information Administration.
Hydrocarbon gas liquids
explained
;
https://www.eia.gov/energyex
plained/hydrocarbon-gas-
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
G
liquids/prices-for-hydrocarbon-gas-liquids.php
(accessed 2022-01-
31).
(6) Wofsy, S. C. HIAPER Pole-to-Pole Observations (HIPPO):
Fine-Grained, Global-Scale Measurements of Climatically Important
Atmospheric Gases and Aerosols.
Philos. Trans. R. Soc. Math. Phys.
Eng. Sci.
2011
,
369
(1943), 2073
2086.
(7) Thompson, C. R.; Wofsy, S. C.; Prather, M. J.; Newman, P. A.;
Hanisco, T. F.; Ryerson, T. B.; Fahey, D. W.; Apel, E. C.; Brock, C.
A.; Brune, W. H.; Froyd, K.; Katich, J. M.; Nicely, J. M.; Peischl, J.;
Ray, E.; Veres, P. R.; Wang, S.; Allen, H. M.; Asher, E.; Bian, H.;
Blake, D.; Bourgeois, I.; Budney, J.; Bui, T. P.; Butler, A.; Campuzano-
Jost, P.; Chang, C.; Chin, M.; Commane, R.; Correa, G.; Crounse, J.
D.; Daube, B.; Dibb, J. E.; DiGangi, J. P.; Diskin, G. S.; Dollner, M.;
Elkins, J. W.; Fiore, A. M.; Flynn, C. M.; Guo, H.; Hall, S. R.; Hannun,
R. A.; Hills, A.; Hintsa, E. J.; Hodzic, A.; Hornbrook, R. S.; Huey, L.
G.; Jimenez, J. L.; Keeling, R. F.; Kim, M. J.; Kupc, A.; Lacey, F.; Lait,
L. R.; Lamarque, J.-F.; Liu, J.; McKain, K.; Meinardi, S.; Miller, D. O.;
Montzka, S. A.; Moore, F. L.; Morgan, E. J.; Murphy, D. M.; Murray,
L. T.; Nault, B. A.; Neuman, J. A.; Nguyen, L.; Gonzalez, Y.; Rollins,
A.; Rosenlof, K.; Sargent, M.; Schill, G.; Schwarz, J. P.; Clair, J. M. St.;
Steenrod, S. D.; Stephens, B. B.; Strahan, S. E.; Strode, S. A.; Sweeney,
C.; Thames, A. B.; Ullmann, K.; Wagner, N.; Weber, R.; Weinzierl, B.;
Wennberg, P. O.; Williamson, C. J.; Wolfe, G. M.; Zeng, L. The
NASA Atmospheric Tomography (ATom) Mission: Imaging the
Chemistry of the Global Atmosphere.
Bull. Am. Meteorol. Soc.
2022
,
103
(3), E761
E790.
(8) Mund, J.; Thoning, K.; Tans, P.; Dlugokencky, E.
NOAA ESRL/
GMD Annual Period of Record Operational Carbon Cycle
&
Greenhouse
Gases (CCGGs) from Surface Stations
2017
,
DOI: 10.7289/
V5CN725S
.
(9) Mund, J.; Thoning, K.; Tans, P.; Andrews, A.; Kofler, J.;
Trudeau, M. E.; Bakwin, P. S.; Fisher, M. L.; Sweeney, C.; Desai, A. R.
Earth System Research Laboratory Carbon Cycle and Greenhouse Gases
Group Continuous Measurements of CO2, CO, and CH4 from Tall
Towers
2017
,
DOI: 10.7289/V57W69F2
.
(10) Mund, J.; Thoning, K.; Tans, P.; Sweeny, C.; Higgs, J.; Wolter,
S.; Crotwell, A.; Neff, D.; Dlugokencky, E.; Lang, P.; Novelli, P.;
Moglia, E.; Crotwell, M.
Earth System Research Laboratory Carbon
Cycle and Greenhouse Gases Group Flask-Air Sample Measurements of
CO2, CH4, CO, N2O, H2, and SF6 from the Aircraft Program
2017
,
DOI: 10.7289/V5N58JMF
.
(11) Vimont, I.; Montzka, S.; Hall, B.; Siso, C.; Crotwell, M.
Preliminary Ethane and Propane Data Used in Tribby et al.
;
https://
gml.noaa.gov/ccgg/arc/?id=155
(accessed 2022-05-27).
(12) Andrews, A. E.; Kofler, J. D.; Trudeau, M. E.; Williams, J. C.;
Neff, D. H.; Masarie, K. A.; Chao, D. Y.; Kitzis, D. R.; Novelli, P. C.;
Zhao, C. L.; Dlugokencky, E. J.; Lang, P. M.; Crotwell, M. J.; Fischer,
M. L.; Parker, M. J.; Lee, J. T.; Baumann, D. D.; Desai, A. R.; Stanier,
C. O.; De Wekker, S. F. J.; Wolfe, D. E.; Munger, J. W.; Tans, P. P.
CO2, CO, and CH4Measurements from Tall Towers in the NOAA
Earth System Research Laboratory
sGlobalGreenhouseGas
Reference Network: Instrumentation, Uncertainty Analysis, and
Recommendations for Future High-Accuracy Greenhouse Gas
Monitoring Efforts.
Atmospheric Meas. Technol.
2014
,
7
(2), 647
687.
(13) Sweeney, C.; Karion, A.; Wolter, S.; Newberger, T.; Guenther,
D.; Higgs, J. A.; Andrews, A. E.; Lang, P. M.; Neff, D.; Dlugokencky,
E.; Miller, J. B.; Montzka, S. A.; Miller, B. R.; Masarie, K. A.; Biraud, S.
C.; Novelli, P. C.; Crotwell, M.; Crotwell, A. M.; Thoning, K.; Tans,
P. P. Seasonal Climatology of CO
2
across North America from
Aircraft Measurements in the NOAA/ESRL Global Greenhouse Gas
Reference Network: VERTICAL CLIMATOLOGY OF CO
2
IN NA.
J. Geophys. Res. Atmospheres
2015
,
120
(10), 5155
5190.
(14) Jacob, D. J.
Introduction to Atmospheric Chemistry
; Princeton
University Press: Princeton, NJ, 1999; pp 52
53.
(15) Seinfeld, J. H.; Pandis, S. N.
Atmospheric Chemistry and Physics:
From Air Pollution to Climate Change
, 3rd ed.; John Wiley & Sons:
Hoboken, NJ, 2016; p 129.
(16) Keppel-Aleks, G.; Wennberg, P. O.; Schneider, T. Sources of
Variations in Total Column Carbon Dioxide.
Atmospheric Chem. Phys.
2011
,
11
(8), 3581
3593.
(17) Keppel-Aleks, G.; Wennberg, P. O.; Washenfelder, R. A.;
Wunch, D.; Schneider, T.; Toon, G. C.; Andres, R. J.; Blavier, J.-F.;
Connor, B.; Davis, K. J.; Desai, A. R.; Messerschmidt, J.; Notholt, J.;
Roehl, C. M.; Sherlock, V.; Stephens, B. B.; Vay, S. A.; Wofsy, S. C.
The Imprint of Surface Fluxes and Transport on Variations in Total
Column Carbon Dioxide.
Biogeosciences
2012
,
9
(3), 875
891.
(18) Wohltmann, I. Integrated Equivalent Latitude as a Proxy for
Dynamical Changes in Ozone Column.
Geophys. Res. Lett.
2005
,
32
(9),
DOI: 10.1029/2005GL022497
.
(19) Parrish, D. D.; Stohl, A.; Forster, C.; Atlas, E. L.; Blake, D. R.;
Goldan, P. D.; Kuster, W. C.; de Gouw, J. A. E
ff
ects of Mixing on
Evolution of Hydrocarbon Ratios in the Troposphere: MIXING
EFFECTS ON NMHC RATIO EVOLUTION.
J. Geophys. Res.
Atmospheres
2007
,
112
(D10),
DOI: 10.1029/2006JD007583
.
(20) Department of Energy.
Lique
fi
ed Natural Gas: Understanding the
Basic Facts
; DOE/FE-0489; United States of America, 2005.
(21) Rutherford, J. S.; Sherwin, E. D.; Ravikumar, A. P.; Heath, G.
A.; Englander, J.; Cooley, D.; Lyon, D.; Omara, M.; Langfitt, Q.;
Brandt, A. R. Closing the Methane Gap in US Oil and Natural Gas
Production Emissions Inventories.
Nat. Commun.
2021
,
12
(1), 4715.
(22) National Oceanic and Atmospheric Administration (NOAA).
The Science behind the Polar Vortex: You Might Want to Put on a
Sweater
, 2021.
(23) Dalsøren, S. B.; Myhre, G.; Hodnebrog, Ø.; Myhre, C. L.;
Stohl, A.; Pisso, I.; Schwietzke, S.; Höglund-Isaksson, L.; Helmig, D.;
Reimann, S.; Sauvage, S.; Schmidbauer, N.; Read, K. A.; Carpenter, L.
J.; Lewis, A. C.; Punjabi, S.; Wallasch, M. Discrepancy between
Simulated and Observed Ethane and Propane Levels Explained by
Underestimated Fossil Emissions.
Nat. Geosci.
2018
,
11
(3), 178
184.
(24) Helmig, D.; Rossabi, S.; Hueber, J.; Tans, P.; Montzka, S. A.;
Masarie, K.; Thoning, K.; Plass-Duelmer, C.; Claude, A.; Carpenter,
L. J.; Lewis, A. C.; Punjabi, S.; Reimann, S.; Vollmer, M. K.;
Steinbrecher, R.; Hannigan, J. W.; Emmons, L. K.; Mahieu, E.;
Franco, B.; Smale, D.; Pozzer, A. Reversal of Global Atmospheric
Ethane and Propane Trends Largely Due to US Oil and Natural Gas
Production.
Nat. Geosci.
2016
,
9
(7), 490
495.
(25) Höglund-Isaksson, L. Bottom-up Simulations of Methane and
Ethane Emissions from Global Oil and Gas Systems 1980 to 2012.
Environ. Res. Lett.
2017
,
12
(2), 024007.
(26) Franco, B.; Mahieu, E.; Emmons, L. K.; Tzompa-Sosa, Z. A.;
Fischer, E. V.; Sudo, K.; Bovy, B.; Conway, S.; Griffin, D.; Hannigan,
J. W.; Strong, K.; Walker, K. A. Evaluating Ethane and Methane
Emissions Associated with the Development of Oil and Natural Gas
Extraction in North America.
Environ. Res. Lett.
2016
,
11
(4), 044010.
(27) Simpson, I. J.; Sulbaek Andersen, M. P.; Meinardi, S.;
Bruhwiler, L.; Blake, N. J.; Helmig, D.; Rowland, F. S.; Blake, D. R.
Long-Term Decline of Global Atmospheric Ethane Concentrations
and Implications for Methane.
Nature
2012
,
488
(7412), 490
494.
(28) International Energy Agency.
IEA Atlas of Energy: Natural Gas
Production
;
http://energyatlas.iea.org/
#
!/tellmap/-1165808390
(ac-
cessed 2022-05-27).
(29) International Energy Agency.
IEA Atlas of Energy: Crude Oil
Production
;
http://energyatlas.iea.org/
#
!/tellmap/-1920537974
(ac-
cessed 2022-05-27).
(30) U.S. Energy Information Administration.
Drilling Productivity
Report for Key Tight Oil and Shale Gas Regions
,
https://www.eia.gov/
petroleum/drilling/
(accessed 2022-05-27).
(31) Worldwide Power Products.
10 Biggest Shale Plays in the US -
New
&
Used Generators, Ends and Engines
; Worldwide Power
Products: Houston, TX.
(32) Saunois, M.; Stavert, A. R.; Poulter, B.; Bousquet, P.; Canadell,
J. G.; Jackson, R. B.; Raymond, P. A.; Dlugokencky, E. J.; Houweling,
S.; Patra, P. K.; Ciais, P. Global Methane Budget 2000
2017.
Earth
Syst. Sci. Data
2020
,
12
, 1561
1623.
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
H
(33) Scarpelli, T. R.; Jacob, D. J.; Maasakkers, J. D.; Sulprizio, M. P.;
Sheng, J.-X.; Rose, K.; Romeo, L.; Worden, J. R.; Janssens-Maenhout,
G. A Global Gridded (0.1
°×
0.1
°
) Inventory of Methane Emissions
from Oil, Gas, and Coal Exploitation Based on National Reports to
the United Nations Framework Convention on Climate Change.
Earth Syst. Sci. Data
2020
,
12
(1), 563
575.
(34) IEA.
Methane Tracker Database: Interactive Database of Country
and Regional Estimates for Methane Emissions and Abatement Options
;
https://www.iea.org/articles/methane-tracker-data-explorer
(accessed
2022-05-27).
(35) Maasakkers, J. D.; Jacob, D. J.; Sulprizio, M. P.; Turner, A. J.;
Weitz, M.; Wirth, T.; Hight, C.; DeFigueiredo, M.; Desai, M.;
Schmeltz, R.; Hockstad, L.; Bloom, A. A.; Bowman, K. W.; Jeong, S.;
Fischer, M. L. Gridded National Inventory of U.S. Methane
Emissions.
Environ. Sci. Technol.
2016
,
50
(23), 13123
13133.
(36) EPA (Environmental Protection Agency).
Global Anthropogenic
Non-CO2 Greenhouse Gas Emissions: 1990
2030
;
https://www.epa.
gov/global-mitigation-non-co2-gr
eenhouse-gases/global-non-co2-
ghg-emissions-1990-2030
(accessed 2022-05-27).
(37) Turner, A. J.; Jacob, D. J.; Wecht, K. J.; Maasakkers, J. D.;
Lundgren, E.; Andrews, A. E.; Biraud, S. C.; Boesch, H.; Bowman, K.
W.; Deutscher, N. M.; Dubey, M. K.; Griffith, D. W. T.; Hase, F.;
Kuze, A.; Notholt, J.; Ohyama, H.; Parker, R.; Payne, V. H.;
Sussmann, R.; Sweeney, C.; Velazco, V. A.; Warneke, T.; Wennberg,
P. O.; Wunch, D. Estimating Global and North American Methane
Emissions with High Spatial Resolution Using GOSAT Satellite Data.
Atmospheric Chem. Phys.
2015
,
15
(12), 7049
7069.
(38) U.S. Energy Information Administration.
Petroleum
&
Other
Liquids: Oklahoma Field Production of Crude Oil
;
https://www.eia.
gov/dnav/pet/hist/LeafHandler.ashx?f=M&=PET&=MCRFPOK1
(accessed 2022-05-27).
(39) U.S. Energy Inform
ation Administration.
Natural Gas:
Oklahoma Natural Gas Marketed Production
;
https://ycharts.com/
indicators/us_natural_gas_
marketed_production_ngm
#
:~:text=
US%20Natural%20Gas%20Marketed%20Production%20is%20a
t%20a%20current%20level,2.586T%20one%20year%20ago
. (accessed
2022-05-27).
(40) U.S. Energy Information Administration.
Oklahoma State Pro
fi
le
Analysis
, 2021.
(41) Crowley, K.; Collins, R. Oil Producers Are Burning Enough
Waste
Gas to Power Every Home in Texas.
Bloomberg
April 10,
2019
;
https://www.bqprime.com/business/permian-basin-is-
fl
aring-
more-gas-than-texas-residents-use-daily
.
(42) Liu, Z.; Ciais, P.; Deng, Z.; Lei, R.; Davis, S. J.; Feng, S.; Zheng,
B.; Cui, D.; Dou, X.; Zhu, B.; Guo, R.; Ke, P.; Sun, T.; Lu, C.; He, P.;
Wang, Y.; Yue, X.; Wang, Y.; Lei, Y.; Zhou, H.; Cai, Z.; Wu, Y.; Guo,
R.; Han, T.; Xue, J.; Boucher, O.; Boucher, E.; Chevallier, F.; Tanaka,
K.; Wei, Y.; Zhong, H.; Kang, C.; Zhang, N.; Chen, B.; Xi, F.; Liu, M.;
Bréon, F.-M.; Lu, Y.; Zhang, Q.; Guan, D.; Gong, P.; Kammen, D. M.;
He, K.; Schellnhuber, H. J. Near-Real-Time Monitoring of Global
CO2 Emissions Reveals the Effects of the COVID-19 Pandemic.
Nat.
Commun.
2020
,
11
(1), 5172.
Environmental Science & Technology
pubs.acs.org/est
Article
https://doi.org/10.1021/acs.est.2c00927
Environ. Sci. Technol.
XXXX, XXX, XXX
XXX
I