1. CH 4 and CO 2 from space : The SCIAMACHY, GOSAT and Precursor S5 missions Guerlet, Schepers, Galli, Butz 1, Frankenberg 2, Hasekamp, Landgraf, Houweling, Ilse Aben SRON Netherlands Institute for Space Research, Utrecht, The Netherlands 1 Institute for Meteorology and Climate research, Karlsruhe Institute Technology, Germany 2 Jet Propulsion Laboratory, USA

2. 2 Global surface network high accuracy, but limited in coverage (e.g. Tropics) to accurately quantify contribution of natural and anthropogenic emissions. Satellite observations are complementary. Less accurate but with global coverage But requirements very demanding (sub) % level

3. 3 Blackbody curves, solar vs. earth radiation SWIR vs TIR measurements (passive) SWIRTIR

4. 4 Blackbody curves, solar vs. earth radiation

5. SWIR CH 4 (and CO 2 ) missions : 5 SCIAMACHY on ENVISAT 2002 – 2013/2014 CH 4 (and CO 2 ) and many other species TROPOMI on Precursor Sentinel-5 2014 – 2021, Sentinel-5 2020 - … CH 4 and many other species GOSAT 2009 - … CH 4 and CO 2 OCO Only CO 2, 2013

6. 66 http://www.jaxa.jp/press/2009/02/20090209_ibuki_e.html#at1

7. 77 GOSAT (CO 2, CH 4 )

8. 88 http://www.jaxa.jp/press/2009/02/20090209_ibuki_e.html#at1 SCIAMACHY (CH 4 ) GOSAT (CO 2, CH 4 )

9. 99 http://www.jaxa.jp/press/2009/02/20090209_ibuki_e.html#at1 SCIAMACHY (CH 4 )

10. 10 http://www.jaxa.jp/press/2009/02/20090209_ibuki_e.html#at1 TROPOMI CH 4 and CO (and H 2 O)

11. 11 http://www.jaxa.jp/press/2009/02/20090209_ibuki_e.html#at1 TROPOMI CH 4 and CO (and H 2 O) SENTINEL-5

12. 12 One of the most critical issues is scattering by aerosols and cirrus (cloud-free observations)

13. Sciamachy CO 2 TOMS AI July: October: Sahara SCIAMACHY CO 2 Houweling et al., ACP (2005) 370 ppm390 ppm

14. 14 SCIAMACHY, GOSAT (1.6 m) GOSAT (0.75,1.6,2,2.1 m), TROPOMI (0.75 & 2.3 m), heritage Sciamachy Neither method is perfect, methods can be tested using GOSAT Frankenberg et al., Science 2005 RemoteC algorithms

15. 2003-2008 SCIAMACHY average Frankenberg, Science 2005

16. Time-series over the Sahara Increase in methane observed as of 2007 Frankenberg, JGR 2011

17. 17 GOSAT : 1 st dedicated GHG mission (2009 - …) Multi-band high spectral resolution FTIR Circular FOV, 10 km Cloud Aerosol Imager High spatial resolution 500 meter

18. 18 Validation columns CH 4 and CO 2 GOSAT : Total Carbon Column Observing Network (TCCON) Ground-based network of FTIR spectrometers set up for the validation of satellite based CO 2 (and CH 4, N 2 O, HF,CO, H 2 O and HDO) column measurements. We used 12 stations for validation of our GOSAT CH 4 and CO 2 columns

19. 19 CO 2 Processed ~ 1,5 years station-to-station bias variability : ~0.23% (stdv) Scatter ~ 1% Butz, GRL 2011

20. 20 station-to-station bias variability : ~0.23% (stdv) Scatter ~ 1% Proxy and Full-Physics show very similar TCCON comparison CH 4 Butz, GRL 2011

21. 21 Number of important corrections applied : -Ad-hoc additive offset correction to correct for non-linearity -O 2 scaling to correct for probably spectroscopy issues -Apply cirrus filter

22. 22 Global data set ~ 1.5 years

23. 23 GOSAT XCH 4 proxy vs full physics: from the comparison with TCCON, the 2 data products have similar quality. Comparison of the global maps over one year : Sahara shows the greater difference: overestimation from full physics, due to high albedo + aerosol load? Remark: those are seasonal biases

24. 24 CH 4 Proxy versus full physics method : -TM5-NOAA always lower -Difference between proxy and full physics GOSAT in spring-summer. Proxy lower. -Spring-summer dust storm periods  overestimation by full physics due to high aerosol load and high albedo Sahara Schepers, 2011

25. 25 India CH4/CO2 CO2 full physics CO2 Carbon Tracker Carbon Tracker not perfect, effect visible in proxy CH4

26. The ESA Sentinel-5 Precursor (S-5P) is a pre-operational mission focussing on global observations of the atmospheric composition for air quality and climate. The TROPOspheric Monitoring Instrument (TROPOMI) is the payload of the S-5P mission and is jointly developed by The Netherlands and ESA. The planned launch date for S-5P is 2014 with a 7 year design lifetime. sentinel-5 precursor GMES ATMOSPHERE MISSION IN POLAR ORBIT ‣ TROPOMI: UV-VIS-NIR-SWIR push- broom grating spectrometer. ‣ Spectral range: 270-500 nm, 675- 775 nm, 2305-2385 nm ‣ Spectral Resolution: 0.25-1.1 nm ‣ Observation Mode: Nadir, global daily coverage, ground pixel 7x7km 2 at nadir ‣ Orbit: Sun synchronous, 824 km, 13:30 hr dayside equator crossing time. SATELLITE PAYLOAD ‣ O 3 : total and tropospheric column, profile ‣ NO 2 : tropospheric and total column ‣ CO: total column ‣ SO 2 : total column ‣ CH 4 : total column ‣ Aerosol: absorbing index, type, optical depth ‣ CH 2 O: tropopsheric column ‣ H 2 O: total column ‣ BrO: total column CONTRIBUTION TO GMES SERVICES

27. 27 CH 4 TROPOMI simulations cirrus/aerosol error : TROPOMI is able to correct – to large extent- for lightpath modification due to cirrus/aerosols, in fact similarly to GOSAT. Other wavelength band, and lower spectral resolution. Butz, Rem.Sens.Environ. 2011

28. Project lead : Michael Buchwitz, Institute of Environmental Physics (IUP), University of Bremen, Bremen, Germany Essential Climate Variable (ECV) Greenhouse Gases (GHG), CO 2 & CH 4 ESA Climate Change Initiative : Provide consistent long term data on ECVs from space

29. GHG-CCI Project Overview Goal: To deliver global atmospheric CO 2 and CH 4 information needed for a better understanding of regional GHG surface fluxes (sources & sinks) Core products: Column-averaged near-surface-sensitive CO 2 and CH 4, i.e., XCO 2 and XCH 4, from SCIAMACHY/ENVISAT & TANSO-FTS/GOSAT; generated with ECV Core Algorithms (ECAs); several ECAs per product in competition; the best algorithm for a given product will be selected after 2 year Round Robin phase with several algorithm-improvement – processing – analysis cycles Additional constraints products: CO 2 and CH 4 profiles and partial columns from AIRS, IASI, ACE-FTS, MIPAS, SCIAMACHY solar occultation; generated with Additional Constraints Algorithms (ACAs) ECV generation: In year 3 using selected best algorithm(s) (+ intermediate products if available) Activities: User requirements definition, algorithm improvements, data processing and analysis, calibration improvements, validation,... Linked to and complementary with European GMES Global Atmospheric Core Service (MACC & follow-ons)

30. 30 summary : SCIAMACHY already provides since 2003 important 1st global view on CH 4 columns from space. Used in global emission inversions. GOSAT 1 st dedicated greenhouse gas mission now providing its 1 st results for CH 4 and CO 2. Validation with TCCON stations shows very promising results : Station-to-station bias variability ~a few tens of a percent, scatter <1% Global data processing on-going, ready for source/sink inversions. GOSAT data is ideally suited to test different retrieval approaches TROPOMI Precursor Sentinel 5 will continue SCIAMACHY and GOSAT CH 4 series as of ~2015, followed by Sentinel 5 from 2020 onwards Space based observations of CH 4 and CO 2 look very promising !!!

31. 31

32. 32 Challenge: accuracy Houweling et al., ACP, 2010 “precisions of 1-2 ppm [out of roughly 380 ppm] are needed on regional scales to improve our knowledge of carbon cycle phenomena” (Miller et al., JGR, 2007) “failure to limit the regional biases to within a few tenths of a part per million would have a detrimental impact on the flux estimation” (Chevallier et al., JGR, 2006)

33. 33 Retrieval method: concept [CO 2 vertical profile and/or CH 4 vertical profile, scattering parameters, surface parameters, instrument parameters] (vector) RTM + parameterization of particle amount, size, type, height Minimize Phillips-Tikhonov cost function: GOSAT / OCO-2 / TROPOMI observations in the SWIR Butz et al., Appl. Opt., 2009; JGR, 2010; RSE, submitted.

34. 34 Retrieval method: forward model Butz et al., Appl. Opt., 2009; JGR, 2010; RSE, submitted. Parameterized height distribution: Gaussian function of center height z s Parameterized particle size distribution: Power-law with size parameter  s Particle amount: N s [particles/cm 2 ]

35. 35 GOSAT: retrieval setup - details Spectroscopy: Hartmann + Tran, line- mixing (O 2, CO 2 ) and collision induced absorption (O 2 ), HITRAN 2008 Meteorology (p, T, H 2 O): ECMWF ERA interim analysis (1.5º x1.5º, 6 hourly) Surface topography: GTOPO30 database Initial guess CO 2, CH 4 : CarbonTracker, TM4 Cloud screening: TANSO-CAI L2 product Measurements: Total radiance (Stokes’ I) from TANSO-FTS L1B product Instrument: Polarization model by A. Kuze, D. O’Brien. Solar model: Geoff Toon's linelist

36. 36 Performance evaluation: “full-physics” method - GOSAT Full physics vs proxy-CO 2 vs non-scattering red blue black

37. 7 years of SCIAMACHY data (2003-2009) Frankenberg et al, JGR 2011 Alternative view: - increase 2007 in CH4 evident - clear variations visible in Tropics (can not be 'seen' by Surface network) - Negative tropical anomaly in 2006 Inverse modeling using satellite and surface network needed to further investigate Need also to further investigate effect radiation damage