GREENHOUSEGASES
Global assessment of oil and gas
methane ultra-emitters
T. Lauvaux^1 *, C. Giron^2 , M. Mazzolini^2 , A. dÕAspremont2,3, R. Duren4,5, D. Cusworth^6 ,
D. Shindell7,8,9, P. Ciais1,10
Methane emissions from oil and gas (O&G) production and transmission represent a considerable
contribution to climate change. These emissions comprise sporadic releases of large amounts of
methane during maintenance operations or equipment failures not accounted for in current inventory
estimates. We collected and analyzed hundreds of very large releases from atmospheric methane
images sampled by the TROPOspheric Monitoring Instrument (TROPOMI) between 2019 and 2020.
Ultra-emitters are primarily detected over the largest O&G basins throughout the world. With a total
contribution equivalent to 8 to 12% (~8 million metric tons of methane per year) of the global
O&G production methane emissions, mitigation of ultra-emitters is largely achievable at low costs
and would lead to robust net benefits in billions of US dollars for the six major O&G-producing
countries when considering societal costs of methane.
A
s the second most important contributor
to global warming, methane (CH 4 ) has
continued to accumulate in the atmo-
sphere at a rate of ~50 million metric
tons (Mt) per year over the past two
decades, primarily because of increases in
agricultural activities, waste management,
coal, and oil and gas (O&G) production ( 1 , 2 ).
Large discrepancies between atmospheric
inversions, bottom-up inventories, and bio-
geochemical models remain largely unexplained
( 1 , 3 – 5 ). This complicates attribution of the
recent global rise in atmospheric methane to
an anthropogenic or biogenic source, a pos-
sible decline in the atmospheric OH radical
sink ( 6 , 7 ), or changes in biogenic or anthro-
pogenic sources ( 8 ). Evidence of a large under-
estimation of fossil sources was suggested
by a recent analysis of^14 CH 4 isotopic ratios
( 9 ). Representing a quarter of anthropogenic
emissions alone, emissions from O&G pro-
duction activities have increased from 65 to
80 Mt per year over the past 20 years ( 10 ).
This rapid increase imperils the success of
the Paris Agreement ( 11 ). Anthropogenic emis-
sions trends are partly explained by the in-
crease in shale gas production in the US,
which will soon be followed by the develop-
ment of large, currently underexploited shale
reserves in China, Africa, and South America
( 12 ). Although O&G emissions from national
inventories have been widely underestimated
by conventional reporting ( 13 ), airborne imag-
ery surveys have confirmed the omnipresence
of intermittent emissions, distributed accord-
ing to a power law ( 14 – 16 ) with a righthand
tail resulting from very large O&G emissions,
often referred to as super-emitters ( 17 )(top1%
of emitters or >25 kg/hour) ( 18 ).
Until recently, observation-based CH 4 emis-
sion quantification efforts were restricted
regionally to short-duration aircraft surveys
(lasting a few weeks) ( 19 ) or the deployment
of in situ sensor networks ( 20 , 21 ). Global
efforts were limited by sparse sampling of
coarse-resolution CH 4 column retrievals, such
as the GOSAT mission ( 22 ). More routine and
higher spatially resolved emission quantifica-
tion was made possible by the European Space
Agency Sentinel 5-P satellite mission, which
carried the TROPOspheric Monitoring In-
strument (TROPOMI; launched 2018) ( 23 ).
TROPOMI samples daily CH 4 column mole
fractions over the whole globe at moderate
resolutions (5.5 km by 7 km^2 ) and has revealed
multiple individual cases of unintended very
large leaks ( 24 ) and regional basin-wide anom-
alies ( 25 , 26 ). We systematically examine this
dataset over multiple locations worldwide,
which allows us to statistically characterize
visible ultra-emitters (>25 tons/hour) of CH 4
from O&G activities across various basins.
By nature, reducing these ultra-emitters by
enforcing leak detection and repair strat-
egies or by reducing venting during routine
maintenance and repairs provides an action-
able and cost-efficient solution for emission
abatement ( 27 ).
Detection of atmospheric column CH 4 en-
hancements from single point sources is limited
by TROPOMI instrument sensitivity [5 to 10parts per billion (ppb)] ( 28 ), by the overlap of
multiple plumes from closely located natural
gas facilities (e.g., in the Permian basin), and by
complex spatial gradients from remote sources
that affect background conditions (supple-
mentary materials). Rapidly varying meteoro-
logical conditions require sufficiently robust
approaches, especially with curved CH 4 plume
structures for which common mass balance
methods are too simplistic ( 29 ). We addressed
this problem by applying an automated plume
detection algorithm and quantified the asso-
ciated emissions using the Lagrangian particle
model HYSPLIT ( 30 ) driven by meteorological
reanalysis products for each detected plume
enhancement (>25 ppb averaged over sev-
eral pixels; supplementary materials) over the
whole globe. The detection threshold was
adjusted to exclusively capture statistically
significant enhancements against highly var-
iable backgrounds (supplementary materials).
Finally, we estimated the potential reductions
along with abatement costs for various coun-
tries, to determine effective gains at national
levels.
The number of detections of large total col-
umn CH 4 mole fraction enhancements around
theworld,eachassociatedwithanultra-emitter,
totals >1800 single observed anomalies over
2 years (2019–2020); a large fraction of them
are located over Russia, Turkmenistan, the US
(excluding the Permian basin where regional
enhancements comprise many small to medium
emitters), the Middle East, and Algeria (Fig. 1).
Detections vary in magnitude and number
(between 50 and 150 per month), most of them
corresponding to O&G production or trans-
mission facilities (about two-thirds of detec-
tions, or ~1200), whereas ultra-emitters from
coal, agriculture, and waste management rep-
resent only a relatively small fraction (33%)
of total detections (supplementary mate-
rials). Ultra-emitters attributed to O&G infra-
structure appear along major transmission
pipelines and over most of the largest O&G
basins, representing more than 50% of total
onshore natural gas production worldwide
( 10 ). Offshore emissions remain invisible to
TROPOMI, and cloud cover almost entirely
blocks O&G basins in tropical areas; hence,
these are excluded from our analysis (supple-
mentary materials).
Estimated emissions from O&G ultra-emitters
rank highest for Turkmenistan with 1.3 Mt
of CH 4 per year, followed by Russia, the US
(excluding the Permian basin), Iran, Kazakhstan,
and Algeria (Fig. 2A). Because leak duration
varies and S5-P provides only snapshots, each
leak duration was determined either on the
basis of an observed duration deduced from
the plume length (advection time) or setting a
24-hour duration when consecutive images can
confirmthepresenceofthesameanomalyover
multiple days (Fig. 2A). Leaks lasting several daysSCIENCEscience.org 4 FEBRUARY 2022•VOL 375 ISSUE 6580 557
(^1) Laboratoire des Sciences du Climat et de lÕEnvironnement,
IPSL, Univ. de Saclay, Saclay, France.^2 Kayrros, 33 rue
Lafayette, 75009 Paris, France.^3 CNRS & DI, Ecole Normale
Supérieure, Paris, France.^4 Office of Research, Innovation
and Impact, University of Arizona, Tucson, AZ, USA.^5 Carbon
Mapper, 12 S. Raymond St., Suite B, Pasadena, CA 91105,
USA.^6 Jet Propulsion Laboratory, California Institute of
Technology, Pasadena, CA, USA.^7 Earth & Climate Sciences
Division, Nicholas School of the Environment, Duke
University, Durham, NC, USA.^8 Porter School of the
Environment and Earth Sciences, Tel Aviv University,
Tel Aviv, Israel.^9 Climate and Clean Air Coalition, 1 Rue
Miollis, Building VII, F-75015 Paris, France.^10 Climate and
Atmosphere Research Centre, the Cyprus Institute
(CyI), Nicosia, 2121, Cyprus.
*Corresponding author. Email: [email protected]
RESEARCH | REPORTS