ACPAtmospheric Chemistry and PhysicsACPAtmos. Chem. Phys.1680-7324Copernicus PublicationsGöttingen, Germany10.5194/acp-16-14091-2016Quantifying the loss of processed natural gas within California's South Coast Air Basin using long-term measurements of ethane and methaneWunchDebradwunch@atmosp.physics.utoronto.cahttps://orcid.org/0000-0002-4924-0377ToonGeoffrey C.HedeliusJacob K.https://orcid.org/0000-0003-2025-7519VizenorNicholasRoehlColeen M.SaadKatherine M.https://orcid.org/0000-0002-2501-6223BlavierJean-François L.BlakeDonald R.WennbergPaul O.https://orcid.org/0000-0002-6126-3854Department of Physics, University of Toronto, Toronto, CanadaDivision of Geological and Planetary Sciences, California Institute of Technology, Pasadena, California, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USADepartment of Chemistry, University of California, Irvine, California, USADivision of Engineering and Applied Science, California Institute of Technology, Pasadena, California, USADebra Wunch (dwunch@atmosp.physics.utoronto.ca)15November20161622140911410526April201617May201613September201618October2016This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://acp.copernicus.org/articles/16/14091/2016/acp-16-14091-2016.htmlThe full text article is available as a PDF file from https://acp.copernicus.org/articles/16/14091/2016/acp-16-14091-2016.pdf
Methane emissions inventories for Southern California's South Coast Air Basin (SoCAB)
have underestimated emissions from atmospheric measurements. To provide
insight into the sources of the discrepancy, we analyze records of
atmospheric trace gas total column abundances in the SoCAB starting in the
late 1980s to produce annual estimates of the ethane emissions from
1989 to 2015 and methane emissions from 2007 to 2015. The first decade of
measurements shows a rapid decline in ethane emissions coincident with
decreasing natural gas and crude oil production in the basin. Between 2010
and 2015, however, ethane emissions have grown gradually from about
13 ± 5 to about 23 ± 3 Gg yr-1, despite the steady
production of natural gas and oil over that time period. The methane
emissions record begins with 1 year of measurements in 2007 and continuous
measurements from 2011 to 2016 and shows little trend over time, with an
average emission rate of 413 ± 86 Gg yr-1. Since 2012, ethane to
methane ratios in the natural gas withdrawn from a storage facility within
the SoCAB have been increasing by 0.62 ± 0.05 % yr-1,
consistent with the ratios measured in the delivered gas. Our atmospheric
measurements also show an increase in these ratios but with a slope of
0.36 ± 0.08 % yr-1, or 58 ± 13 % of the slope
calculated from the withdrawn gas. From this, we infer that more than half of
the excess methane in the SoCAB between 2012 and 2015 is attributable to losses
from the natural gas infrastructure.
Introduction
Anthropogenic sources of the potent greenhouse gas methane (CH4)
constitute about 60 % of the global total CH4 emissions, or
nearly 350 Tg CH4 yr-1. Urban regions are
thought to be an important contributor to this flux
e.g.,, and thus both quantification and attribution of these
urban sources are crucial for fully understanding their causes and hence
potentially regulating them. Southern California's South Coast Air Basin
(SoCAB) has been the focus of several studies. These studies have quantified
the emissions from the basin and generally find that the SoCAB emissions are
higher than the reported inventories
.
The SoCAB is a highly urbanized region centred on Los Angeles, with almost
17 million residents, representing 43 % of the population of California.
The lower atmosphere over the SoCAB is well confined: it is contained by
mountains to the north and east and open to the Pacific Ocean to the
south-west. Thus, urban emissions within the basin have long residence times
and, under prevailing wind conditions, also have strong and predictable
diurnal flow: out to the ocean at night and inland during the day.
The many sources of methane in the SoCAB include oil and gas exploration and
extraction, natural gas delivery pipelines and storage facilities,
wastewater treatment plants, landfills, and dairies. Previous studies have
shown that the atmosphere over the SoCAB contains significant CH4
enhancements over the global background .
More recent work has attempted to attribute the sources of the enhanced
methane using other tracers in the atmosphere that are co-emitted with
particular sources. used simultaneous measurements of
ethane (C2H6) and methane to separate ethane-containing sources of
methane, such as natural gas and petroleum, from biogenic sources of methane
which do not co-emit ethane, such as landfills, wastewater treatment, and
ruminants. inferred that a significant fraction of the
excess methane in the SoCAB atmosphere is likely emitted from the natural gas
infrastructure, potentially post-consumer metre. used
co-emitted higher-order alkanes (including ethane) to suggest that oil and
gas drilling and storage are significant contributors to the elevated methane
and ethane emissions. and
conclude that most of the elevated methane in the western SoCAB is related to
fossil fuels using spatial alkane measurements and isotope measurements,
respectively.
We describe our data records and analysis methodology in
Sect. , and in Sect. we discuss the change
in the emissions of methane and ethane within the SoCAB. By comparing the
ethane to methane ratios measured in the atmosphere with the changing ratios
in the withdrawn and delivered natural gas, we quantify the fraction of the
excess methane in the atmosphere attributable to the natural gas
infrastructure.
Methods
We use data from four solar viewing ground-based Fourier transform
spectrometers (FTS) that have measured within the SoCAB. The first
instrument, the JPL MkIV FTS , has measured ethane,
methane, and other trace gases from the Jet Propulsion Laboratory (JPL, NASA)
since 1985 (Fig. ). The measurements have been made
once or twice per week, for about 2 h per day, when the instrument is not in
the field elsewhere for intensive scientific campaigns. Two other instruments
were temporarily stationed at JPL: JPL2007
was operational between July 2007 and June 2008, and JPL2011
was operational between July 2011 and July 2013. These
instruments measured CH4 and other gases, but not C2H6, and
are part of the Total Carbon Column Observing Network TCCON;
. The fourth instrument, which is located about 10 km
from JPL at the California Institute of Technology (Caltech), is part of the
TCCON and has been measuring ethane, methane, and other trace gases with high
temporal frequency (several hundred spectra per sunny day) since September
2012 . The JPL MkIV FTS data are available from
the MkIV website (http://mark4sun.jpl.nasa.gov/ground.html), and
the TCCON data are available from the TCCON archive
(http://tccon.ornl.gov/).
Time series from the MkIV FTS in the SoCAB. The colourful diamonds
are the background surface in situ values measured atop Mauna Loa. The black
circles indicate the MkIV FTS measurements of XCO (top),
XCH4 (middle), and XC2H6 (bottom). There is a marked
decrease in both the day-to-day variability and median value in
XCO over time, an increase in XCH4 in line with the
global trends, and non-monotonic, seasonal changes in
XC2H6.
Both the MkIV and TCCON FTS instruments are direct solar viewing and measure
solar absorption by atmospheric trace gases; the retrievals are thus
insensitive to atmospheric aerosol abundances. The data analysis for these
instruments makes use of the GGG2014 software package . This
includes a nonlinear least squares spectral fitting algorithm (GFIT) that
scales an a priori profile for best fit and a spectroscopic line list
based on the HITRAN database . The
GGG2014 software produces column-averaged dry-air mole fractions of the trace
gas of interest (Xgas), which is defined as
Xgas=columngascolumnairdry.
The column of dry air, in units of molecules cm-2, is computed either
from retrieved oxygen (O2), when available (for the TCCON records), or
from precise measurements of the surface pressure (for the MkIV
record):
columnairdry=columnO20.2095=Ps{g}airmairdry-columnH2OmH2Omairdry.
The measured surface pressure (Ps) is converted to a dry surface
pressure by subtracting the column amount of water
(columnH2O), where {g}air is the column-averaged
gravitational acceleration, mairdry is the molecular mass
of dry air, and mH2O is the molecular mass of water.
The MkIV time series plots shown in Fig. reflect
the influence of local sources in addition to the large-scale backgrounds for
these gases. To show the global background trends, overlaid on
Fig. are the surface in situ measurements of
methane , carbon monoxide CO;
, and ethane made atop Mauna Loa, Hawaii.
The apparent “noise” in the MkIV time series is both from diurnal changes
and from the larger seasonal changes. Note that the magnitude of the Mauna
Loa free-tropospheric in situ concentrations should not be expected to exactly
match the MkIV total column-averaged dry-air mole fractions. In particular,
the concentration of methane is significantly lower above the tropopause, and
so XCH4 is generally lower than the free-tropospheric methane
concentrations .
To diagnose the contribution of SoCAB sources to the trace gas columns, we
quantify the diurnally varying gas ratios following the methodology described
in detail in and briefly described as follows. Because of
the topography of the SoCAB and its predictable diurnal wind flow pattern,
gases emitted into the basin atmosphere, even if they are not emitted by the
same source, show similar diurnal patterns, with a peak in the total column
around 14:00 local time, when the planetary boundary layer is thickest.
Diurnal changes thus represent emissions into the SoCAB. To quantify the
diurnal change in Xgas for the TCCON data, we subtract morning
values from afternoon values at the same solar zenith angles, producing
ΔXgas, a “gas anomaly” value. This approach minimizes
air-mass-dependent biases in the
measurements from appearing as diurnal changes, but it does not remove the
small temperature bias (as afternoons are systematically warmer than
mornings). However, sensitivity studies which perturb the assumed lower
atmosphere temperature show that the temperature bias has a small effect on
the diurnal change of the trace gases, with magnitudes of ≲5 %
of the total diurnal variability (see Appendix ).
We assume that the emissions into the lowest layers of the atmosphere cause
the diurnal pattern in Xgas and thus we explicitly account for
differences in the measurement sensitivity at the surface to each gas by
dividing the ΔXgas by the value of the column averaging
kernel at the surface. We then compute the slope that relates anomalies of
one gas to another. Our data filtering scheme, designed to minimize the
impacts of non-basin air, fires, significant weather events, and instrument
problems, is described in Appendix .
The MkIV dataset is temporally sparse, and the observation strategy was not
intended for this kind of differential analysis: MkIV measurements are taken
around solar noon, and only for 1 to 2 h per day. While this observation
strategy minimizes air-mass variation, columns measured only an hour apart
tend to be similar, and so the computed anomalies are small and therefore
noisy. A consequence of this is that MkIV methane measurements, which have
smaller fractional diurnal variability than the other gases presented here,
are not currently precise enough for anomaly analysis. Daily anomalies of
ethane, carbon monoxide, and acetylene are computed here by subtracting the
daily mean value from each measurement and applying the column averaging
kernel in the same manner as for the TCCON datasets. We aggregate MkIV
ΔXgas data for each year to calculate tracer–tracer
anomaly slopes. Because the TCCON datasets are much denser, we aggregate
monthly data. Subsampling the TCCON datasets to match the times of the MkIV
measurements does not appear to bias the results (see
Appendix ).
To determine emissions of the gas of interest, we use tracer–tracer anomaly
slopes to carbon monoxide, whose emissions in the SoCAB are well
constrained by extensive, biannual, mandatory vehicle smog checks and
oversight by the California Air Resources Board (CARB) and are published
through the CARB web page by air basin
(http://www.arb.ca.gov/app/emsinv/emssumcat.php).
suggested that using CO instead of CO2 to compute emissions
may underestimate the emissions due to different diurnal emissions patterns,
but subsequent studies have shown better agreement with the CH4
emissions estimates computed using its relationship with CO. To calculate the emissions of the
gas of interest, we apply the following equation:
EgasSoCAB=αgasMgasMCOECOSoCAB.
where ECOSoCAB is the emission of carbon monoxide in the
SoCAB in units of TgCO, αgas is the slope of the
correlation between the gas of interest and carbon monoxide in
molmol-1, and Mgas and MCO are the molecular
masses of the gas of interest and carbon monoxide, respectively, in
gmol-1.
The uncertainty estimates on the tracer–tracer anomaly slopes are the
standard deviation of many slopes calculated by bootstrapping
a linear fit that takes x and y errors into account
. Uncertainty estimates on the emissions are determined by
multiplying the calculated emissions by the sum in quadrature of the
fractional uncertainties of the slopes and the assumed uncertainty on the
CARB carbon monoxide emissions (20 %).
Ancillary data
To determine the composition of the natural gas delivered to the SoCAB, we
collected bi-weekly samples of the natural gas delivered to Caltech by
SoCalGas. Natural gas components were separated using gas chromatography on
an HP-PLOT Q column. The abundance of each gas was measured using a flame
ionization detector with appropriate calibrations. To ensure no drift in the
chromatograph, a natural gas standard was also regularly analyzed. Prior to
November 2014 the analysis was performed on site on the same day the sample
was collected. Afterwards, samples were collected in canisters and analyzed
in batches using an off-site gas chromatograph, also using a PLOT column and
flame ionization detector.
To determine the composition of the natural gas stored within the SoCAB, we
use data made publicly available by the Southern California Gas Company
(SoCalGas). There are four SoCalGas gas storage facilities (Aliso Canyon in
Northridge, Honor Rancho in Valencia, Golita near Santa Barbara, and Playa
Del Rey), two of which are within the SoCAB (Aliso Canyon and Playa Del Rey).
Both the Aliso Canyon and Playa Del Rey facilities are exhausted oil wells
that were re-purposed to store natural gas. The Aliso Canyon facility is one
of the largest depleted-well gas storage facilities in the United States,
with an 168 billion cubic foot capacity (4.8 billion cubic metres)
; the Playa Del Rey facility can store only about 2
billion cubic feet (∼ 1 % of the Aliso Canyon capacity). As the
result of a 2007 legal settlement, SoCalGas publishes monthly withdrawn gas
composition from the Playa Del Rey wells . The data are
freely obtained from their website
(https://www.socalgas.com/stay-safe/pipeline-and-storage-safety/playa-del-rey-storage-operations).
The Aliso Canyon facility does not regularly make their withdrawn gas
composition publicly available. However, between October 2015 and February
2016, they made daily atmospheric measurements near the facility available on
their website in response to the failure of one of the withdrawal wells that
resulted in a large loss of gas
(https://www.alisoupdates.com/acu-aliso-canyon-air-sample-results).
Other measurements from aircraft near the facility have been recently
published .
Defining local plumes within the data
Highly local plumes of methane are periodically observed throughout the
Caltech FTS time record. We define these “plumes” as a diurnal change in
methane that is not correlated with an associated change in carbon monoxide.
Carbon monoxide is a heavily emitted gas within the SoCAB, but it has no
significant common sources with methane, so correlations between carbon
monoxide and methane are due to the SoCAB's atmospheric dynamics and thus
represents what we will refer to as the “ambient” SoCAB air.
To quantify this, we use quantile–quantile plots that
determine whether two datasets draw from the same probability distribution.
In these plots, a linear relationship indicates that the distributions are
similar, and any deviations from linearity suggest that the distributions are
different. We assume that the data in the linear region of the graph sample
ambient SoCAB air, and the nonlinear regions are from the plumes.
Figure shows the quantile–quantile plots for
anomalies in methane and carbon monoxide.
These quantile–quantile plots show the extent to which Δ
XCH4 and ΔXCO anomaly data from the Caltech FTS
are from the same probability distribution. When the distributions of the two
datasets are similar, the points (blue “+”) fall along the red dashed
line. The top panel shows the quantile–quantile plot of methane and carbon
monoxide from data prior to the Aliso Canyon gas leak, which started on
23 October 2015. The plot is linear between the grey lines, which indicate
the 95 % quantiles of ΔXCO and ΔXCH4. We use these limits to define air that is representative
of “ambient” SoCAB air from air that contains plumes during that time
period. The bottom panel shows the quantiles of ΔXCH4
and ΔXCO anomaly data for the time period after the Aliso
Canyon gas leak began. For this time period, 80 % quantiles were chosen
to distinguish between ambient and plume air.
From these plots, we determine the regions of nonlinearity, marked by grey
bars. We assume that the data that fall outside the grey bars represent air
that is not well mixed (i.e., “plume” air) and that the “ambient” air is
contained in the box defined by the grey bars. The top panel shows the data
prior to 22 October 2015 and after 11 February 2016, and the bottom panel
shows the data between those dates, during the period of sustained Aliso
Canyon losses.
Results and discussion
We have computed emissions estimates of C2H6 since 1989
(Figs. , )
and CH4 emissions estimates since 2007
(Fig. ). The emissions of ethane in the
basin decreased significantly from the late 1980s
(Fig. ) from 70 ± 17 to
13 ± 5 Ggyr-1 in 2010. These 2010 emissions values agree
well with previous studies (12.9 Gg, ;
11.4 ± 1.6 Gg, ). Since 2010, however,
ethane emissions have nearly doubled. Emissions of CH4 are steady
over the 2007–2016 period, with an average value of
413 ± 86 Ggyr-1 and a slope of
-5 ± 4 Ggyr-1 (-1.2 ± 1.0 %yr-1),
in good agreement with the results from , who have monitored
CH4 in various locations throughout the SoCAB since 2011.
The top panel right axis shows the estimated carbon monoxide
emissions inventory for the SoCAB, published by the California Air Resources
Board (CARB). The top panel left axis shows the inferred emissions of ethane
from the MkIV FTS (black circles), the Caltech FTS (blue squares), previous
estimates from and (green squares), and
(pink diamond). The second panel shows the ethane to
carbon monoxide anomaly slopes from the MkIV FTS (black circles), the Caltech
FTS (blue squares), and previous studies (green squares). The gold line with
gold stars represents what the ethane to carbon monoxide anomaly slope would
be if ethane in the atmosphere remained constant at 1 % of the year 2000
carbon monoxide emissions from 2000 onward. The third panel shows the
acetylene to carbon monoxide anomaly slopes, which are reasonably invariant
over the time series.
There are three main sources of ethane emissions in the SoCAB: vehicle
exhaust, the natural gas system, and oil and gas exploration and extraction.
Of these sources, only vehicle exhaust is not a significant source of
CH4. To distinguish between vehicle exhaust and fossil fuel sources,
we use our coincident measurements of carbon monoxide, which tracks sources
of incomplete combustion (including mobile sources), and acetylene
(C2H2), whose emissions more directly track vehicle exhaust
. The ratio of ethane to
carbon monoxide in the SoCAB declined rapidly until the mid-1990s, and then
slowly and steadily increased. The ratio of acetylene to carbon monoxide
remained relatively constant throughout the time period
(Figs. , ),
and thus the ethane to acetylene ratios follow the same trend as ethane to
carbon monoxide. This implies that vehicle emissions are not driving the
changes in ethane emissions. This is consistent with the
analysis, which showed an increase in ethane relative to acetylene after
1995, which they attributed to natural gas use and production. Using the
motor vehicle gas composition measured by , and the
reported SoCAB carbon monoxide emissions for 1995 by CARB for mobile sources
2.114 Tgyr-1, , we infer that ethane
emissions from mobile sources account for only ∼ 8 % of the
observed ethane, in agreement with the 5–10 % estimate of
for the year 2010. Thus, emissions from vehicles are
unlikely to be either a dominant source of ethane to the SoCAB atmosphere or
responsible for the significant decrease in ethane after 1995. Prior to 1995,
there were fewer regulatory controls on air pollution from vehicles, and the
exhaust composition is much less well known .
This plot shows the monthly methane (top), ethane (middle), and
acetylene (bottom) emissions measured in the atmosphere by the Caltech FTS
(blue squares). Grey solid lines indicate the best-fit slopes with standard
errors indicated by the grey dashed
lines.
Natural gas and crude oil production from the Los Angeles Basin decreased
by about a factor of 2 between 1990 and 2000
. The region's natural gas
liquids production, which includes ethane, propane, and higher-order alkanes,
is negligibly small and no production is reported after 1993
. The Los Angeles Basin and the SoCAB are not
identical regions: the Los Angeles Basin encompasses the SoCAB except for the
north-western corner of Los Angeles County, but it additionally includes the
eastern portions of San Bernardino and Riverside counties and all of San
Diego and Imperial counties. We assume that the production in the SoCAB
tracks the Los Angeles Basin production. The fractional decrease in natural
gas and crude oil production is consistent with the drop in ethane emissions
measured by the MkIV FTS between 1990 and 2000
(Fig. ). However, the absolute abundance is
inconsistent with the 17 % losses from oil and gas extraction determined
by for 2010: it would account for less than half of the
C2H6 emissions in 1990. This suggests that either extraction losses
from oil and gas production in the 1990s were significantly higher or
the ethane content of the gas was larger.
The left-hand axis shows methane emissions measured in the
atmosphere by three TCCON FTS instruments that were located in the SoCAB
since 2007. The grey solid line indicates the best-fit slope with standard
errors indicated by the grey dashed lines. Previous measured emissions are
indicated by green squares. The right-hand axis shows the delivered natural
gas to the SoCAB and is scaled such that if 2 % of the delivered gas is
released into the atmosphere, the atmospheric burden would be equal to the
numbers (in Gg) on the left-hand axis.
Between 2000 and 2010, the ethane emissions remained relatively constant
(Fig. ), consistent with the steady production of
gas and oil. After 2010, however, the calculated ethane emissions increase
monotonically, in contrast with the near-constant oil and gas production.
To explain the ethane increases in the latter period, we rely on our
temporally denser atmospheric measurements from the Caltech FTS, combined
with measurements of ethane and methane available from the withdrawn natural
gas composition of the Playa Del Rey storage facility, and measurements of
the delivered natural gas composition to Caltech.
Figure shows the time series of ethane to methane
ratios since late 2009 from the Playa Del Rey storage facility. The ratios
were roughly constant at around 2.3 % until a minimum in spring 2012 of
∼ 1.7 %. Since that time, the ethane to methane ratios have
increased at a rate of 0.62 ± 0.05 % yr-1 with ratios
exceeding 4 % by mid-2015. This significant increase in ethane content of
the natural gas provides an unique opportunity to attribute the sources of
CH4 to the SoCAB atmosphere. Our measurements of the ethane to
methane ratio in the natural gas delivered to Caltech show values consistent
with the stored natural gas at Playa Del Rey and at Aliso Canyon and a
consistent change in ratio over time (0.59 ± 0.10 % yr-1).
The variability of the ratios measured in the delivered gas is much higher
than that reported by SoCalGas (Fig. ) and
commensurate with the variability seen in the atmospheric measurements. Since
Caltech and Playa Del Rey are located ∼ 45 km apart, this
suggests that the Playa Del Rey withdrawn gas values provide a reasonable (if
smoothed) approximation of the basin-wide natural gas ratios.
Natural gas, crude oil, and natural gas liquids production in the
Los Angeles Basin, reported by
, are shown
in the top panel. The natural gas liquids production values are multiplied by
5 for scale. In the lower panel, the production is scaled to illustrate the
changes in production relative to 2000. The MkIV C2H6 emissions
relative to 2000 (black circles) are added for
reference.
Emissions inventories for CH4 and C2H6. Only one
methane emission value is included, which is the mean emissions over the
2007–2015 measurement period. Ethane emissions are from the Caltech
measurements only, and each column of the table contains data from September
through August. Emissions marked with an asterisk (*) are from
for 2010. The pipeline natural gas emissions of ethane
are computed by multiplying the methane emissions from the pipeline natural
gas (58 % of the measured total CH4) by the increasing slope
fitted to the ethane to methane ratios measured by the Caltech instrument.
Uncertainties on the “measured” emissions are the standard deviations of
the monthly emissions computed for the time range.
Measurements of the atmospheric ethane to methane emissions ratios using the
Caltech FTS data increase by 0.36 ± 0.08 % yr-1, which is
58 ± 13 % of the change in the ratio of ethane to methane reported
in the storage gas by SoCalGas at the Playa Del Rey storage facility. The
linear relationship between the Caltech FTS ethane to methane ratios and the
Playa Del Rey ratios has a slope of 58 ± 12 %
(Fig. ), providing confirmation of this value. This
finding is consistent with more than half of the excess atmospheric burden of
methane in the western SoCAB being attributable to emissions from the natural
gas infrastructure.
This time series shows the ethane to methane ratios in the Playa Del
Rey gas storage facility (brown diamonds), in the natural gas delivered to
the laboratory (grey diamonds), and in gas anomalies measured with the Caltech
FTS (blue squares). The slope of the Playa Del Rey ratios is shown in brown;
the slope of the Caltech FTS ratios is in blue with dashed lines indicating
the slope uncertainty. The slope of the delivered gas samples is not shown,
but it is statistically indistinguishable from the Playa Del Rey slope. The
median ethane to methane anomaly ratio measured by SoCalGas in the air near
the Aliso Canyon gas leak is indicated by the orange triangle, and the value
near Aliso Canyon measured from an aircraft platform by is
indicated by the orange diamond.
Since the average total methane emissions in the SoCAB since 2007 have been
roughly constant at 413 ± 86 Ggyr-1
(Fig. ; Table ), the
∼ 58 % attributable to the natural gas infrastructure is
240 ± 78 Ggyr-1. In 2015, the SoCalGas total throughput
was 2559 MMcfday-1, or 18 TgCH4 total
. We remove 3 TgCH4 from wholesales and
0.2 TgCH4 for company use and “lost and unaccounted for” (LUAF)
gas, giving 14.7 TgCH4 delivered by SoCalGas. This suggests
1.6 ± 0.5 % losses as fugitive emissions from the total delivered.
(However, only 74 % of the population served by SoCalGas lives in the
SoCAB, and thus the fraction of the losses as fugitive emissions would
represent a larger fraction of the delivered gas to SoCAB customers;
.) The roughly constant total CH4 emissions and
delivered natural gas implies that downstream natural gas emissions were not
likely changing during this period. The remaining
∼ 173 ± 56 Ggyr-1 excess methane is likely from
sources lacking an ethane signature that tracks the pipeline natural gas
composition. These likely sources are the SoCAB dairies
, feedlots and range cattle, landfills, septic
systems , and – likely particularly important in
the western part of the basin – oil and gas extraction.
estimate 182 ± 54 GgCH4yr-1 emitted from
methane-dominant sources (i.e., dairies, landfills, and wastewater treatment
plants) and the oil and gas extraction to be
32 ± 7 GgCH4yr-1. Our results are consistent with these
previous studies within the uncertainties. Table compiles
these emissions for CH4 between 2007 and 2015 and for C2H6
between 2012 and 2015. We assume constant total emissions of CH4
during the 2007–2015 period and changing C2H6 emissions from the
increasing ethane content in the pipeline-quality natural gas. Within the
uncertainties, the increase in observed C2H6 emissions can be wholly
explained by the increasing ethane content in the delivered natural gas. The
other sources of C2H6 (vehicular exhaust, oil and gas exploration
and production) are assumed to be constant.
This figure shows the ethane to methane ratios from the Caltech FTS
data on the y axis and from the Playa Del Rey gas storage facility on the
x axis between September 2012 and March 2016. The colours indicate the date
of the measurements. The slope of the relationship is indicated by the black
line (0.58±0.12) and is consistent with the slope derived from
Fig. .
Droughts such as the one plaguing Southern California since 2012/2013
(; ) can reduce the
ability of soil microbes to remove methane and ethane released underground
into the soils (;
). The constant CH4 emissions and growing
C2H6 emissions since 2012 would require a compensating decrease in
biogenic emissions of CH4 to offset this effect. However, biogenic
emissions are reported to have decreased by about 1 % between 2012 and
2014 , so this effect is likely to be small.
Aliso Canyon
A large gas loss from the Aliso Canyon Storage Facility to the SoCAB began on
23 October 2015 according to SoCalGas and reports from those living nearby.
The failed well was finally plugged on 11 February 2016.
estimate that approximately 97.1 GgCH4 were released into the
atmosphere during the 112-day leak, about 25 % of the typical annual
SoCAB methane emissions. After 23 October 2015, we see several days with very
large enhancements in atmospheric methane and ethane, typically in the
afternoons when the plume is advected into the line of sight of the
instruments. We see no evidence of such large plumes prior to 23 October in
our measurements. The plumes from Aliso Canyon can be easily distinguished
from the ambient SoCAB air during this period (Fig. ,
lower panel), and in these plumes the ethane and methane anomalies are very
well correlated with a slope of 4.28 ± 0.07 %
(Fig. ), in good agreement with the
recent delivered natural gas ethane to methane ratios which exceed 4 %.
From our atmospheric measurements and the CH4
emissions estimate, we calculate that the ethane emission from this leak is
7.7 ± 1.7 GgC2H6, which is about 40 % of the annual
SoCAB ethane emissions. estimated a consistent
7.3 GgC2H6 emissions using aircraft measurements.
This figure shows the ethane and methane anomalies during the
Caltech TCCON record. The entire time series is represented by filled
circles, the plume data are represented by “x” symbols, and the measurements
of the plume originating from the Aliso Canyon gas leak are circled in black.
The colours represent the year during which the measurements were recorded.
The average ambient slopes from Fig. are indicated
with solid lines and show a time dependence consistent with the slopes from
plumes. The ethane to methane slope in the Aliso Canyon plume data (black
line) shows a high degree of correlation (R2=0.95) and a slope of
4.28±0.07 %. Note that the ethane to methane ratios in the ambient
air were rising throughout the
record.
While dramatic and important to prevent, the Aliso Canyon well failure
represents only a small fraction of the SoCAB methane emissions over the long
term (< 3 % of the emissions from the SoCAB between 2007 and 2015).
Furthermore, the annual methane emissions into the SoCAB
(10.3 ± 2.2 TgCO2eyr-1, using the 100-year global
warming potential of 25) represent less than 7 % of those of carbon
dioxide (CO2), which we estimate to be 167.4 Tgyr-1 by
scaling the California Air Resources Board estimate for California's carbon
dioxide emissions in 2013 386.6 Tgyr-1; to
the population of the SoCAB. Thus, significantly reducing the long-term
climate impact of the SoCAB's greenhouse gas emissions requires focusing
efforts to reduce carbon dioxide emissions directly.
Conclusions
We have measured the total column atmospheric abundances of ethane, methane,
and other trace gases since the late 1980s in the South Coast Air Basin in
Southern California, USA. We calculate that ethane emissions declined rapidly
until the mid-1990s, coincident with the decline in Los Angeles Basin
production of natural gas and crude oil, but the absolute abundances are
inconsistent with recent estimates of natural gas emissions from the SoCAB
oil and gas production. This may suggest that either extraction losses were
higher in the 1990s than they are today or the ethane content of the
gas was larger. After the mid-1990s, the ethane emissions are relatively
constant until ∼ 2010 and then roughly double between 2010 and 2015.
This increase cannot be explained by the (decreasing) vehicular emissions or
(steady) natural gas and oil production in the basin, but they can be explained by
the increasing ethane content of the natural gas delivered to the SoCAB.
Methane emissions have remained steady since 2007 at
413 ± 86 Ggyr-1. Since 2012, ethane to methane ratios in
the stored and delivered natural gas have increased and are tracked in our
atmospheric measurements with a slope of about 58 ± 13 % the
magnitude, implying that over half of the excess methane in the basin air is
from losses in the natural gas infrastructure. These long-term measurements
allow us to monitor the atmospheric composition and attribute changes in the
atmosphere to specific sources within the basin with unique time
dependencies.
The Aliso Canyon Gas Storage facility well failure on 23 October 2015 was
one of the biggest singular natural gas releases in US history. Our
measurements indicate that this leak, which is estimated by
to have released 97.1 GgCH4 into the SoCAB
atmosphere in just 112 days, produced 7.7 ± 1.7 GgC2H6,
about 40 % of the typical annual ethane emissions in the basin. The
long-term climate impacts from the Aliso Canyon well failure are much smaller
than the accumulated background methane emissions and minor compared with
the direct carbon dioxide emissions in the SoCAB.
Data availability
TCCON data are available from the TCCON data archive, hosted by CDIAC:
http://tccon.ornl.gov. Each TCCON dataset used in this paper is cited
independently. The JPL MkIV FTS data are available from the webpage
http://mark4sun.jpl.nasa.gov/ground.html.
Data filtering
Data from the Caltech FTS (N=73 335) were filtered to avoid biases in the
slopes using the following criteria:
There must be at least five measurements during the day to calculate ΔXgas anomalies.
We filter out days on which the ΔXCO2 changes by less than
1.5 ppm, as those are typically days during which the prevailing
winds are so-called “Santa Anas”, which bring relatively clean air from the
Mohave Desert from the north into the SoCAB and hence are not representative
of SoCAB air.
We filter out days on which hydrogen fluoride anomalies (ΔXHF)
change by more than 10 ppt. ΔXHF is a proxy for
tropopause height, and large changes in it over the course of the day
indicates a front or other significant weather change not representative of
typical SoCAB air.
We filter out days on which the biomass burning tracer ΔXHCN
changes by more than 0.5 ppt, because these data are likely
contaminated with fire emissions.
Each month of data must contain at least 15 ΔXgas points
for a slope to be calculated for that month. This avoids biasing the slopes
based on a few non-representative measurements.
Ethane and methane are measured on two separate detectors: ethane is measured
with an InSb detector and methane with an InGaAs detector. Both detectors
measure carbon monoxide, and so we ensure that the carbon monoxide measured
on the two detectors are consistent. Any measurements for which the carbon
monoxide in the two bands differ by more than 2σ of their mean
difference are excluded from further analysis.
Data from the MkIV FTS were filtered more loosely (N=1727) than the Caltech
FTS measurements, as the density of measurements is much lower, and
measurements are manually initiated and terminated within a few hours of noon
on clear, smoke-free days.
There must be at least 5 ΔXgas anomalies per year to
calculate the tracer–tracer slopes.
The change in XCO must be sufficiently large
(5×1017moleculescm2, or ∼ 2 %) in order to
calculate a robust slope for each year.
Temperature bias
The GGG2014 analysis software uses a single a priori temperature profile
throughout each day that is representative of the local noon temperature
profile, derived from the NCEP/NCAR reanalysis data . On
sunny days, there is a systematic increase in surface temperature throughout
the day in the SoCAB, typically a 5 K difference between mid-morning
and mid-afternoon at the surface (see Figs. and
); temperature changes aloft should be
smaller and thus the integrated temperature error throughout the planetary
boundary layer should be smaller than 5 K.
To minimize the temperature sensitivity of our retrievals, we chose windows
in which the target absorption lines have average ground-state energies of
around 300 cm-1. For example, we use the entire CO and
CH4 bands in the near infrared, which have roughly the same number of
high-j and low-j lines, reducing the temperature sensitivity.
C2H6 is measured in its Q branches between 2976 and
2997 cm-1. Based on performing C2H6 retrievals using
correct and incorrect (perturbed) temperature profiles under a range of
different temperature and humidity conditions, we have determined that the
retrieved C2H6 amount will change by less than 1 % for a
temperature perturbation of 5 K at the surface, decreasing to zero at
3.5 km altitude. Since a typical diurnal change between mid-afternoon
and mid-morning in the retrieved C2H6 is about 20 %, the
temperature-induced affect is comparatively small. A similar sensitivity
study for CH4 resulted in errors of less than 0.02 % for surface
temperature changes on the order of 18 K. This is significantly
smaller than the < 1 % diurnal variations in CH4 in the SoCAB.
This figure shows the change in surface temperature throughout the
day from the noontime a priori value (top panel) and the diurnal surface
temperature error in the bottom panel. The diurnal surface temperature error
is computed by subtracting morning from afternoon surface temperatures in the
same way as the trace gas anomalies are
computed.
This figure shows a histogram of the surface temperature anomalies
shown in the bottom panel of
Fig. .
Sampling bias
The sampling strategies of the MkIV and Caltech FTS measurements differ
significantly. MkIV observations are performed manually. From JPL, the MkIV
measures within 1 h of solar noon, once or twice per week. The Caltech
instrument is automated and measures throughout the day, every day, whenever
it is sunny. To determine whether biases caused by these sampling differences
affect the results of our analyses, we selected a subset of the coincident
time series from MkIV and Caltech in 2015. We then filtered both datasets
according to Appendix and further subselected the
Caltech data to points within 15 min of the MkIV measurements. The grey dots
in Fig. show all the filtered Caltech data; black
diamonds are the Caltech data time-matched with MkIV; red circles are the
MkIV data themselves. Slopes of the tracer–tracer anomalies in the third
panel below show a small bias between the filtered and time-matched Caltech
slopes, both well within the uncertainties of the MkIV slope. Thus, there
should not be a significant bias introduced into the tracer–tracer slopes
from the sampling strategy.
This figure shows the negligible impact on the derived tracer–tracer
slope from the different sampling strategies used by the MkIV and Caltech
measurements. The top panel shows the time series for CO total column
abundances for Caltech (grey), MkIV (red), and the subsampled Caltech values
to coincide with the MkIV measurements (black). The middle panel shows the
C2H6 time series. The bottom panel shows the tracer–tracer
relationship between the diurnal anomalies of the trace gases. The slopes
computed for the three cases agree well within
error.
Acknowledgements
Part of this work was performed at the Jet Propulsion Laboratory, California
Institute of Technology, under contract with NASA. We thank the various
people who have assisted with MkIV ground-based observations over the years,
as well as the NASA Upper Atmosphere Research Program for funding. This
research was supported by NASA's Carbon Cycle Science program (NNX14AI60G).
TCCON data were obtained from the TCCON Data Archive, hosted by the Carbon
Dioxide Information Analysis Center (CDIAC) – http://tccon.ornl.gov.
We thank four anonymous reviewers and D. Lyon for providing thoughtful
reviews that significantly improved this paper. Edited by: R.
Volkamer Reviewed by: four anonymous referees
ReferencesAdamse, A. D., Hoeks, J., Bont, J. A. M., and Kessel, J. F.: Microbial
activities in soil near natural gas leaks, Arch. Mikrobiol., 83,
32–51, 10.1007/BF00425043, 1972.AQMD: Aliso Canyon Update, Tech. rep., South Coast Air Quality Management
District, available at: http://www.aqmd.gov/home/regulations/compliance/aliso-canyon-update (last access: 24 February 2016),
2016.California Gas and Electric Utilities: 2016 California Gas Report, Tech.
rep., available at: https://www.socalgas.com/regulatory/cgr.shtml (last access: 10 November 2016), 2016.CARB: Almanac Emission Projection Data, available at:
http://www.arb.ca.gov/app/emsinv/emssumcat.php (last access:
22 March 2016), 2009.CARB: Greenhouse Gas Emissions Inventory Summary [2000–2013], Tech.
rep.,
California Air Resources Board, Sacramento, available at:
http://www.arb.ca.gov/app/ghg/2000_2013/ghg_sector_data.php (last
access: 24 February 2016),
2015.CARB: Greenhouse Gas Emission Inventory Summary [2000–2014], Tech. rep.,
California Air Resources Board, Sacramento, available at:
http://www.arb.ca.gov/app/ghg/2000_2014/ghg_sector_data.php (last
access: 26 August 2016),
2016.Conley, S., Franco, G., Faloona, I., Blake, D. R., Peischl, J., and Ryerson,
T. B.: Methane emissions from the 2015 Aliso Canyon blowout in Los Angeles,
CA, Science, 2348, 2–7, 10.1126/science.aaf2348, 2016.Crounse, J. D., DeCarlo, P. F., Blake, D. R., Emmons, L. K., Campos, T. L.,
Apel, E. C., Clarke, A. D., Weinheimer, A. J., McCabe, D. C., Yokelson, R.
J., Jimenez, J. L., and Wennberg, P. O.: Biomass burning and urban air
pollution over the Central Mexican Plateau, Atmos. Chem. Phys., 9,
4929–4944, 10.5194/acp-9-4929-2009, 2009.Dlugokencky, E. J., Crotwell, A., Lang, P., and Mund, J.: Atmospheric
Methane
Dry Air Mole Fractions from quasi-continuous measurements at Mauna Loa,
Hawaii, 1986–2015, Version: 2016-01-20, available at:
ftp://aftp.cmdl.noaa.gov/data/trace_gases/ch4/in-situ/surface/ (last
access: 4 February 2016),
2016.Efron, B. and Gong, G.: A Leisurely Look at the Bootstrap, the Jackknife,
and
Cross-Validation, Am. Stat., 37, 36–48, 10.2307/2685844, 1983.Griffin, D. and Anchukaitis, K. J.: How unusual is the 2012–2014
California drought?, Geophys. Res. Lett., 41, 9017–9023,
10.1002/2014GL062433, 2014.Helmig, D., Hueber, J., and Tans, P. P.: Non- Methane Hydrocarbons from the
NOAA ESRL Surface Network, 2004–2015, available at:
ftp://aftp.cmdl.noaa.gov/data/trace_gases/voc/c2h6/flask/surface/c2h6_mlo_surface-flask_1_arl_event.txt
(last access: 4 February 2016),
2015.Hopkins, F. M., Kort, E. A., Bush, S. E., Ehleringer, J. R., Lai, C.-T.,
Blake,
D. R., and Randerson, J. T.: Spatial patterns and source attribution of
urban methane in the Los Angeles Basin, J. Geophys. Res.-Atmos., 121, 2490–2507, 10.1002/2015JD024429, 2016.Hsu, Y.-K., VanCuren, T., Park, S., Jakober, C., Herner, J., FitzGibbon, M.,
Blake, D. R., and Parrish, D. D.: Methane emissions inventory verification
in southern California, Atmos. Environ., 44, 1–7,
10.1016/j.atmosenv.2009.10.002,
2010.Kirchstetter, T. W., Singer, B. C., Harley, R. A., Kendall, G. R., and Chan,
W.: Impact of Oxygenated Gasoline Use on California Light-Duty Vehicle
Emissions, Environ. Sci. Technol., 30, 661–670,
10.1021/es950406p, 1996.McKain, K., Wofsy, S. C., Nehrkorn, T., Eluszkiewicz, J., Ehleringer, J. R.,
and Stephens, B. B.: Assessment of ground-based atmospheric observations for
verification of greenhouse gas emissions from an urban region, P. Natl. Acad. Sci. USA, 109,
8423–8428, 10.1073/pnas.1116645109,
2012.Novelli, P. and Masarie, K.: Atmospheric Carbon Monoxide Dry Air Mole
Fractions from the NOAA ESRL Carbon Cycle Cooperative Global Air Sampling
Network, 1988-2014, Version: 2015-08-04,
available at:
ftp://aftp.cmdl.noaa.gov/data/trace_gases/co/flask/surface/ (last
access: 4 February 2016),
2015.Peischl, J., Ryerson, T. B., Brioude, J., Aikin, K. C., Andrews, A. E.,
Atlas,
E., Blake, D. R., Daube, B. C., de Gouw, J. A., Dlugokencky, E. J., Frost,
G. J., Gentner, D. R., Gilman, J. B., Goldstein, A., Harley, R. A., Holloway,
J. S., Kofler, J., Kuster, W. C., Lang, P. M., Novelli, P. C., Santoni,
G. W., Trainer, M., Wofsy, S. C., and Parrish, D. D.: Quantifying sources of
methane using light alkanes in the Los Angeles basin, California, J.
Geophys. Res.-Atmos., 118, 4974–4990, 10.1002/jgrd.50413, 2013.Rothman, L., Gordon, I., Babikov, Y., Barbe, A., Chris Benner, D., Bernath,
P. F., Birk, M., Bizzocchi, L., Boudon, V., Brown, L. R., Campargue, A.,
Chance, K., Cohen, E., Coudert, L., Devi, V., Drouin, B., Fayt, A., Flaud,
J.-M., Gamache, R., Harrison, J., Hartmann, J.-M., Hill, C., Hodges, J.,
Jacquemart, D., Jolly, A., Lamouroux, J., Le Roy, R., Li, G., Long, D.,
Lyulin, O., Mackie, C., Massie, S., Mikhailenko, S., Müller, H.,
Naumenko, O., Nikitin, A., Orphal, J., Perevalov, V., Perrin, A., Polovtseva,
E., Richard, C., Smith, M., Starikova, E., Sung, K., Tashkun, S., Tennyson,
J., Toon, G. C., Tyuterev, V., and Wagner, G.: The HITRAN2012 molecular
spectroscopic database, J. Quant. Spectrosc. Ra., 130, 4–50, 10.1016/j.jqsrt.2013.07.002,
2013.Saad, K. M., Wunch, D., Toon, G. C., Bernath, P., Boone, C., Connor, B.,
Deutscher, N. M., Griffith, D. W. T., Kivi, R., Notholt, J., Roehl, C.,
Schneider, M., Sherlock, V., and Wennberg, P. O.: Derivation of tropospheric
methane from TCCON CH4 and HF total column observations, Atmos. Meas.
Tech., 7, 2907–2918, 10.5194/amt-7-2907-2014, 2014.Saunois, M., Bousquet, P., Poulter, B., Peregon, A., Ciais, P., Canadell, J.
G., Dlugokencky, E. J., Etiope, G., Bastviken, D., Houweling, S.,
Janssens-Maenhout, G., Tubiello, F. N., Castaldi, S., Jackson, R. B., Alexe,
M., Arora, V. K., Beerling, D. J., Bergamaschi, P., Blake, D. R., Brailsford,
G., Brovkin, V., Bruhwiler, L., Crevoisier, C., Crill, P., Curry, C.,
Frankenberg, C., Gedney, N., Höglund-Isaksson, L., Ishizawa, M., Ito, A.,
Joos, F., Kim, H.-S., Kleinen, T., Krummel, P., Lamarque, J.-F., Langenfelds,
R., Locatelli, R., Machida, T., Maksyutov, S., McDonald, K. C., Marshall, J.,
Melton, J. R., Morino, I., O'Doherty, S., Parmentier, F.-J. W., Patra, P. K.,
Peng, C., Peng, S., Peters, G. P., Pison, I., Prigent, C., Prinn, R.,
Ramonet, M., Riley, W. J., Saito, M., Schroeder, R., Simpson, I. J., Spahni,
R., Steele, P., Takizawa, A., Thornton, B. F., Tian, H., Tohjima, Y., Viovy,
N., Voulgarakis, A., van Weele, M., van der Werf, G., Weiss, R., Wiedinmyer,
C., Wilton, D. J., Wiltshire, A., Worthy, D., Wunch, D. B., Xu, X., Yoshida,
Y., Zhang, B., Zhang, Z., and Zhu, Q.: The Global Methane Budget: 2000–2012,
Earth Syst. Sci. Data Discuss., 10.5194/essd-2016-25, in review, 2016.SoCalGas: Playa del Rey natural gas storage operations, Tech. rep.,
Southern California Gas Company, available at:
https://www2.socalgas.com/documents/safety/pdr_storage.pdf (last
access: 10 November 2016),
2008.
Swain, D., Tsiang, M., Haugen, M., Singh, D., Charland, A., Rajaratnam, B.,
and
Diffenbaugh, N. S.: The extraordinary California drought of 2013/2014:
Character, context, and the role of climate change, B. Am. Meteorol. Soc.,
95, S3–S7, 2014.Toon, G. C.: The JPL MkIV interferometer, Optics and Photonics News, 2,
19–21,
10.1364/OPN.2.10.000019, 1991.Toon, G. C.: Telluric line list for GGG2014,
10.14291/tccon.ggg2014.atm.R0/1221656, 2014.Toon, G. C. and Blavier, J.-F. L.: MkIV Ground-based Vertical Column
Abundances 1985–2015, available at:
http://mark4sun.jpl.nasa.gov/ground.html (last access:
10 November 2016), 2015.Townsend-Small, A., Tyler, S. C., Pataki, D. E., Xu, X., and Christensen,
L. E.: Isotopic measurements of atmospheric methane in Los Angeles,
California, USA: Influence of “fugitive” fossil fuel emissions, J. Geophys. Res., 117, 1–11, 10.1029/2011JD016826,
2012.USEIA: California – Los Angeles Basin Onshore Crude Oil Estimated
Production
from Reserves (Million Barrels), available at:
http://tonto.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=RCRR10RCAL_1&f=A
(last access: 16 August 2016),
2015a.USEIA: California – Los Angeles Basin Onshore Natural Gas Plant Liquids,
Reserves
Based Production (Million Barrels),
available at: http://tonto.eia.gov/dnav/ng/hist/rl2r55rcal_1a.htm (last
access: 16 August 2016),
2015b.USEIA: California – Los Angeles Basin Onshore Dry Natural Gas Reserves
Estimated Production (Billion Cubic Feet), available at:
http://tonto.eia.gov/dnav/ng/hist/rngr20rcal_1a.htm (last access:
16 August 2016),
2015c.USEIA: Natural Gas Annual Respondent Query System (EIA-191 Field Level
Storage Data (Annual) through 2014), available at:
http://www.eia.gov/cfapps/ngqs/ngqs.cfm?f_report=RP7 (last access:
25 February 2016),
2016.van den Pol-van Dasselaar, A., van Beusichem, M. L., and Oenema, O.: Effects
of soil moisture content and temperature on methane uptake by grasslands on
sandy soils, Plant Soil, 204, 213–222, 10.1023/A:1004371309361, 1998.Viatte, C., Lauvaux, T., Hedelius, J. K., Parker, H., Chen, J., Jones, T.,
Franklin, J. E., Deng, A. J., Gaudet, B., Verhulst, K., Duren, R., Wunch, D.,
Roehl, C., Dubey, M. K., Wofsy, S., and Wennberg, P. O.: Methane emissions
from dairies in the Los Angeles Basin, Atmos. Chem. Phys. Discuss.,
10.5194/acp-2016-281, in review, 2016.Wang, Z., Deutscher, N. M., Warneke, T., Notholt, J., Dils, B., Griffith, D.
W. T., Schmidt, M., Ramonet, M., and Gerbig, C.: Retrieval of tropospheric
column-averaged CH4 mole fraction by solar absorption FTIR-spectrometry
using N2O as a proxy, Atmos. Meas. Tech., 7, 3295–3305,
10.5194/amt-7-3295-2014, 2014.Warneke, C., de Gouw, J. A., Holloway, J. S., Peischl, J., Ryerson, T. B.,
Atlas, E., Blake, D. R., Trainer, M., and Parrish, D. D.: Multiyear trends
in volatile organic compounds in Los Angeles, California: Five decades of
decreasing emissions, J. Geophys. Res.-Atmos., 117,
1–10, 10.1029/2012JD017899, 2012.Washenfelder, R. A., Wennberg, P. O., and Toon, G. C.: Tropospheric methane
retrieved from ground-based near-IR solar absorption spectra, Geophys.
Res. Lett., 30, 1–5, 10.1029/2003GL017969,
2003.Wennberg, P. O., Mui, W., Wunch, D., Kort, E. A., Blake, D. R., Atlas, E. L.,
Santoni, G. W., Wofsy, S. C., Diskin, G. S., Jeong, S., and Fischer, M. L.:
On the sources of methane to the Los Angeles atmosphere, Environ.
Sci. Technol., 46, 9282–9289, 10.1021/es301138y, 2012.Wennberg, P. O., Roehl, C., Blavier, J.-F., Wunch, D., Landeros, J., and
Allen,
N.: TCCON data from Jet Propulsion Laboratory, Pasadena, California, USA,
Release GGG2014R0, 10.14291/tccon.ggg2014.jpl02.R0/1149297, 2014a.Wennberg, P. O., Wunch, D., Roehl, C., Blavier, J.-F. L., Toon, G. C., and
Allen, N.: TCCON data from California Institute of Technology, Pasadena,
California, USA, Release GGG2014R1,
10.14291/tccon.ggg2014.pasadena01.R1/1182415, 2014b.Wennberg, P. O., Wunch, D., Yavin, Y., Toon, G. C., Blavier, J.-F., Allen,
N., and Keppel-Aleks, G.: TCCON data from Jet Propulsion Laboratory,
Pasadena, California, USA, Release GGG2014R0,
10.14291/tccon.ggg2014.jpl01.R0/1149163, 2014c.Wilk, M. B. and Gnanadesikan, R.: Probability plotting methods for the
analysis of data, Biometrika, 55, 1–17, 10.1093/biomet/55.1.1,
1968.Wong, K. W., Fu, D., Pongetti, T. J., Newman, S., Kort, E. A., Duren, R.,
Hsu, Y.-K., Miller, C. E., Yung, Y. L., and Sander, S. P.: Mapping
CH4 : CO2 ratios in Los Angeles with CLARS-FTS from Mount Wilson,
California, Atmos. Chem. Phys., 15, 241–252, 10.5194/acp-15-241-2015,
2015.Wong, C. K., Pongetti, T. J., Oda, T., Rao, P., Gurney, K. R., Newman, S.,
Duren, R. M., Miller, C. E., Yung, Y. L., and Sander, S. P.: Monthly trends
of methane emissions in Los Angeles from 2011 to 2015 inferred by CLARS-FTS
observations, Atmos. Chem. Phys., 16, 13121–13130,
10.5194/acp-16-13121-2016, 2016.Wunch, D., Wennberg, P. O., Toon, G. C., Keppel-Aleks, G., and Yavin, Y. G.:
Emissions of greenhouse gases from a North American megacity, Geophys.
Res. Lett., 36, 1–5, 10.1029/2009GL039825,
2009.Wunch, D., Toon, G. C., Blavier, J.-F. L., Washenfelder, R. A., Notholt, J.,
Connor, B. J., Griffith, D. W. T., Sherlock, V., and Wennberg, P. O.: The
total carbon column observing network, Philos. T. R. Soc. A, 369, 2087–2112,
10.1098/rsta.2010.0240, 2011.Wunch, D., Toon, G. C., Sherlock, V., Deutscher, N. M., Liu, C., Feist,
D. G.,
and Wennberg, P. O.: The Total Carbon Column Observing Network's GGG2014
Data Version, Tech. rep., California Institute of Technology, Carbon Dioxide
Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge,
Tennessee, USA,
10.14291/tccon.ggg2014.documentation.R0/1221662,
2015.
York, D., Evensen, N. M., Martinez, M. L., and De Basabe Delgado, J.:
Unified
equations for the slope, intercept, and standard errors of the best straight
line, Am. J. Phys., 72, 367, 10.1119/1.1632486,
2004.