A view on mid-tropospheric CH4 in the tropics:
6 years from MetOp-A/IASI
C. Crevoisier1, R. Armante1, S. Massart2, C. Cressot3, L. Crépeau1,
N. Meilhac1, O. Membrive1, N. A. Scott1, and A. Chédin1
1Laboratoire
de Météorologie Dynamique/CNRS/IPSL 2ECMWF 3LSCE/IPSL
Contact: cyril.crevoisier@lmd.polytechnique.fr
Introduction
Methane (CH4) is the third most important greenhouse gas in the atmosphere, a7er water vapor and CO2, and plays a significant role in tropospheric chemistry through its removal of OH. There is thus a considerable interest in monitoring its evoluDon to beEer characterize its atmospheric budget and infer how it might evolve in the future. The rate of increase in CH4 has been steadily decreasing since the late 1980s, reaching a near-­‐zero increase from 1999 to 2006. From the end of 2006 or beginning of 2007, a renewed growth of atmospheric CH4 has been observed by surface networks (Rigby et al., 2009; Dlugokencky et al., 2009), potenDally stemming from a combinaDon of a slight change in OH, and increased emissions in northern high and low laDtude regions. Here, we present 6.5 years of mid-­‐tropospheric methane derived from the Infrared Atmospheric Sounding Interferometer (IASI) over the tropics, spanning July 2007-­‐Dec. 2013, from which we infer characterisDcs of tropospheric methane and its most recent evoluDon. The
retrieval
method
Use is made of a non linear inference scheme based on neural networks (Crevoisier et al., 2009), iniDally designed for the retrieval of CO2 (Chédin et al., 2003). The IASI instrument
IASI, developed by CNES in collaboraDon with EUMETSAT, is a Fourier Transform Spectrometer based on a Michelson Interferometer coupled to an integrated imaging system that measures infrared radiaDon emiEed from the Earth. • Launched on 19th October 2006 onboard MetOp-­‐A and declared operaDonal in July 2007, IASI provides: -­‐8461 spectral channels between 645 and 2760 cm−1 (15.5 -­‐ 3.63 μm). -­‐with a spectral resoluDon of 0.5 cm−1 a7er apodisaDon. -­‐the spectral sampling interval is 0.25 cm−1. • IASI channels are sensiDve to methane around 7.7 µm as can be seen in the figure below that shows the variaDon of Brightness Temperature induced on each IASI channel by a typical variaDon of various atmospheric gas concentraDons. SelecDon of a set of CH4 channels Training data set (TIGR atmospheric database) nl =1
ΔTB (qgas ) =
Off-­‐line radiaDve bias removal nl =1
• IASI data are rouDnely archived at LMD via the Ether Centre for Atmospheric Chemistry Products and Services (hEp://ether.ipsl.jussieu.fr/), through EUMETCast, the Broadcast System for Environmental Data of EUMETSAT. Year-to-year evolution
IASI shows an increase of mid-­‐tropospheric CH4 from 2007 to 2009, larger in the southern than in the northern tropics for 2007 and 2008, in agreement with the renewed growth of methane measured at the surface since the end of 2006 as shown by the GLOBALVIEW-­‐CH4 network. CH
(ppbv) CH
4 4(ppbv) Zonally averaged tch4
ropical dfonction
istribuDon f m
id-­‐tropospheric CH a4 s Zonally averaged ropical of o[200708->201311]
m
id-­‐tropospheric CH
Cycle t
du
(ppbv) d
enistribuDon de la latitude
4
as rfrom etrieved from IASI ffrom 2011 2013 retrieved Metop-­‐A/IASI rom JJuly uly 22007 007 tto o FNeb. ovember 1780
1860
1770
1760
1840
1750
1820
1740
1800
1730
1780
1720
1710
1760
1700
1740
1780
1770
1860
1760
1840
1750
1820
1740
1800
1730
1720
1780
1710
1760
1700
1740
20
20
15
15
10
10
55
-5
-5
1001
1201
0907
1101
-10
-10
0901
1001
-15
0807
0901
-15
month
month
-20 08010801
-20
0707
1007
1301
Month Month • Seasonal cycle: -­‐more pronounced in the northern (~25 ppbv) than in the southern (~15 ppbv) tropics. -­‐In the northern tropics: maximum in November and minimum in March-­‐April. -­‐In the southern tropics: more complex seasonal variaDon due to year-­‐to-­‐year fluctuaDons of UT wind distribuDon at ~200 hPa which modulates the inter-­‐hemispheric transport during ENSO phases (Prinn et al., 1992), yielding enhanced exchanges in December-­‐May during cold phases (as in 2007 and 2008), and a cancellaDon in the southern tropics of the summer minimum due to photochemical destrucDon of methane. CARIBIC • Surface fluxes es\ma\on (August 2009-­‐July2010): • Use of LSCE Bayesian inversion system. • Use of surface network (NOAA), TANSO-­‐FTS (UoL) and IASI (LMD) CH4. • Good staDsDcal consistency between CH4 flux esDmates from each datasets. H1
H2
H3
H4
H5
HIPPO CH4 (ppbv)
1780
1760
1740
Over 13 CARIBIC flights, on a 4°x4° grid: IASI – CARIBIC = 7.2 ± 13.1 ppbv 1740
1760
1780
IASI CH4 (ppbv)
1800
1820
1840
Over the 5 HIPPO campaigns: IASI – HIPPO = 5.1 ± 16.3 ppbv •  Chédin et al. (2003) First global measurement of midtropospheric CO2 from NOAA polar satellites: Tropical zone, J. Geophys. Res., 108, 4581 Ÿ Cressot et al. (2014) On the consistency between global and regional methane emissions inferred from SCIAMACHY, TANSO-­‐FTS, IASI and surface measurements. Atmos. Chem. Phys., 14, 577-­‐592. ŸCrevoisier et al. (2009) Tropospheric methane in the tropics -­‐ first year from IASI hyperspectral infrared observaDons, Atmos. Chem. Phys., 9, 6337-­‐6350. Ÿ Crevoisier et al. (2013) The 2007–2011 evoluDon of tropical methane in the mid-­‐troposphere as seen from space by MetOp-­‐A/IASI, Atmos. Chem. Phys. Dlugokencky et al. (2009) ObservaDonal constraints on recent increases in the atmospheric CH4 burden, Geophys. Res. Le$., 36, L18803 Ÿ Massart et al. (2013) AssimilaDon of atmospheric methane products in the MACC-­‐II system: from SCIAMACHY to TANSO and IASI. Atmos. Chem. Phys. Discuss. 13, 8023-­‐806413 Ÿ Matsueda and Inoue, (1996) Measurements of atmospheric CO2 and CH4 using a commercial airliner from 1993 to 1994, Atmospheric Environment, 30, 1647-­‐1655 Ÿ Prinn et al. (1992) Global Average ConcentraDon and Trend for Hydroxyl Radicals Deduced From ALE/GAGE Trichloroethane (Methyl Chloroform) Data for 1978–1990, J. Geophys. Res., 97(D2), 2445-­‐2461 Ÿ Rigby et al. (2008) Renewed growth of atmospheric methane, Geophys. Res. Le$., 35, L22805 GLOBALVIEW-­‐CH4 2007 2008 2009 2010 2011 • MACC monitoring and assimila\on: • IASI CH4 is assimilated within ECMWF/IFS since Dec. 2010 (Massart et al., 2014). • In terms of column, the addiDon of the IASI data to TANSO data has a significant impact, mainly over extratropical regions, reducing the global bias. 1800
1720
1720
IASI Use of IASI CH4
1840
1820
CH4 instantaneous growth rate (ppbv yr-­‐1) IASI has the poten\al to follow methane trend on the long-­‐term. HIPPO CONTRAIL 2006-­‐2009 IASI 2007-­‐ Fit of IASI CH4 (whole tropical band) This evoluDon is consistent with the increase of CH4 emission from tropical wetlands due to enhanced La Niña precipitaDon in 2007 and 2008. IASI data suggests a slowing down of the increase in 2009, and a decrease in 2010. This evoluDon is in agreement with the evoluDon of precipitaDon in the tropics and the absence of significant change in fires, and thus in biomass burning CH4 emission, in the tropical region over 2007-­‐2009. • Valida\on with aircra^ measurements: CONTRAIL Chédin et al., 1985 ara.abct.lmd.polytechnique.fr According to the CH4 Jacobians, which are ploEed on the le7 hand side, IASI channels are mostly sensiDve to CH4 in the mid-­‐troposphere. Retrieval of a mid-­‐tropospheric integrated content of CH4. CH4 mid-­‐
tropospheric content 6.5 years of mid-tropospheric CH4
latitude
LaDtude latitude
LaDtude Jacquinet-­‐Husson et al., 2008 ether.ipsl.jussieu.fr €
• The retrieval is limited to the tropical zone because of the greater stability of the tropospheric temperature verDcal which yields a beEer precision. 00
for a given gas €
• Simultaneous use of IR (IASI) channels sensiDve to T and CH4 and of MW (AMSU also flying onboard MetOp-­‐A) channels only sensiDve to T allows to decorrelate T/CH4. CH4
∑
∂TB
(nl) * Δqgas (nl)
∂qgas
Sco$ et Chédin, 1981 www.novelDs.fr/4AOP/ Non-­‐linear inference scheme « clear sky » detecDon ∂T
nl =42
Training of Neural Networks IASI & AMSU BT For each channel and each TIGR situaDon, the variaDon of BT (in K) for a given variaDon of one atmospheric variable is computed using the 4AOP-­‐2009 Jacobians: nl =42
∂TB
for temperature ΔTB (T) = ∑
(nl) * ΔT(nl)
T (1K) H2O (20%) N2O (2%) CH4 (10%) Tsurf (1 K) CH4 global emissions (Ÿ) and losses (⏏) Although IASI is mostly sensi\ve to the mid-­‐troposphere, it does provide informa\on on surface fluxes. 549±36 538±15 565±21 Cressot et al., ACP, 2014 PRIOR TANSO-­‐FTS IASI SURFACE Conclusion
CH4 fields retrieved from IASI seem to confirm that tropical wetland emissions driven by higher than average precipitaDon are one of the main drivers of the 2007-­‐2008 increase. The 2010 decrease of methane suggested by IASI is also supported by the drop in precipitaDon over this year in wetland tropical region. The results presented here show that IASI can provide a powerful constraint on the monitoring and understanding of atmospheric methane burden. With the launch of two other successive IASI-­‐like instruments in 2012 and 2018, more than 20 years of observaDons of mid-­‐tropospheric methane will be available for climate studies.