EMMA CO2
A joint effort to deliver satellite retrieved atmospheric CO2 concentrations for surface flux inversions The ensemble median algorithm EMMA
Overview Since long time, climate modellers use ensemble approaches to calculate the ensemble median and to estimate uncertainties of climate projections where no ground-truth is known. Following this idea, the ensemble median algorithm EMMA composes level 2 data of up to seven GOSAT and SCIAMACHY XCO2 retrievals independently developed by NASA (ACOS), NIES (operational product, PPDF), SRON (RemoteC), University of Leicester (UOL-FP), and the University of Bremen (BESD, WFMD). EMMA determines in 10x10 degree grid boxes monthly averages and selects the level 2 data of the median algorithm. Intelligent thresholds (depending on potential information content) prevent from over-weighting individual algorithms with a considerably larger amount of data. Contact EMMA, University of Bremen Max Reuter (maximilian.reuter at iup.physik.uni-bremen.de) Institute of Environmental Physics (IUP) University of Bremen, FB1 Otto-Hahn-Allee 1 D-28334 Bremen Germany Phone: +49 (421) 218 62085 FAX: +49 (421) 218 4555 http://www.iup.uni-bremen.de/~mreuter NASA OCO-2 and GOSAT retrievals Chris O'Dell (odell at atmos.colostate.edu) FOCAL and BESD retrieval, University of Bremen Max Reuter (maximilian.reuter at iup.physik.uni-bremen.de) NIES operational retrieval Yoshida Yukio (yoshida.yukio at nies.go.jp) NIES PPDF-S retrieval Andrey Bril (andrey.bril at gmail.com) RemoTeC retrieval, SRON, KIT Andre Butz (andre.butz at dlr.de) Otto Hasekamp (o.p.hasekamp at sron.nl) UoL-FP retrieval, University of Leicester Hartmut Bösch (hb100 at leicester.ac.uk) WFMD retrieval, University of Bremen Oliver Schneising (schneising at iup.physik.uni-bremen.de) + Literature + Image Galery + Version History Data Access The data access is password-restricted. Please use the following form to obtain username and password. Email: Name: Institute: Country: Please tell us in a few words (>3) what you plan with the data: I agree with the conditions of use: After registration, please use the following link to access the data archive: http://www.iup.uni-bremen.de/~mreuter/emma_data/ Till now, the EMMA CO2 data sets cover the period 01/2003-12/2018. The data-format is NCDF. Condition of Use When using EMMA data, you agree ... ... to inform us prior to any publication where EMMA data products are planned to be used, ... to offer us co-authorship for any planned peer-reviewed publication based on EMMA data products (for non peer-reviewed publications it is sufficient if you add an appropriate acknowledgement), ... not to distribute the EMMA data products to any third party (the only exception being colleagues working in your institute, in this case you agree to inform them about the conditions listed here and that they also have to accept these conditions). Acknowledgements EMMA is in part funded by ESA/ESRIN (GHG CCI, GHG CCI-II, ADVANSE), EU (MACC-II, MACC, CityZen, C3S), DLR (SADOS), and the state and the University of Bremen as well as participating institues and funding organizations.

EMMA CO2

A joint effort to deliver satellite retrieved atmospheric CO2 concentrations for surface flux inversions
The ensemble median algorithm EMMA

Overview

Since long time, climate modellers use ensemble approaches to calculate the ensemble median and to estimate uncertainties of climate projections where no ground-truth is known. Following this idea, the ensemble median algorithm EMMA composes level 2 data of up to seven GOSAT and SCIAMACHY XCO2 retrievals independently developed by NASA (ACOS), NIES (operational product, PPDF), SRON (RemoteC), University of Leicester (UOL-FP), and the University of Bremen (BESD, WFMD). EMMA determines in 10x10 degree grid boxes monthly averages and selects the level 2 data of the median algorithm. Intelligent thresholds (depending on potential information content) prevent from over-weighting individual algorithms with a considerably larger amount of data.

Contact

EMMA, University of Bremen
Max Reuter (maximilian.reuter at iup.physik.uni-bremen.de)
Institute of Environmental Physics (IUP)
University of Bremen, FB1
Otto-Hahn-Allee 1
D-28334 Bremen
Germany
Phone: +49 (421) 218 62085
FAX: +49 (421) 218 4555
http://www.iup.uni-bremen.de/~mreuter

NASA OCO-2 and GOSAT retrievals
Chris O'Dell (odell at atmos.colostate.edu)

FOCAL and BESD retrieval, University of Bremen
Max Reuter (maximilian.reuter at iup.physik.uni-bremen.de)

NIES operational retrieval
Yoshida Yukio (yoshida.yukio at nies.go.jp)

NIES PPDF-S retrieval
Andrey Bril (andrey.bril at gmail.com)

RemoTeC retrieval, SRON, KIT
Andre Butz (andre.butz at dlr.de)
Otto Hasekamp (o.p.hasekamp at sron.nl)

UoL-FP retrieval, University of Leicester
Hartmut Bösch (hb100 at leicester.ac.uk)

WFMD retrieval, University of Bremen
Oliver Schneising (schneising at iup.physik.uni-bremen.de)

+ Literature

+ Image Galery

+ Version History

v4.1.CO2 (02.07.2019)
Algorithms: SCIAMACHY BESD v02.01.02, GOSAT ACOS v7.3.10a, GOSAT NIES v02.75BC, GOSAT PPDF-S v02.xx, GOSAT RemoteC v2.3.8, GOSAT UOL-FP v7.2, OCO-2 FOCAL v08, OCO-2 NASA v9.0.03
Minimum number of algorithms used for median calculation: variable, typically number of algorithms minus two.
Remarks: Based on v3.1 but with new retrieval versions and OCO-2 data sets, bug fixes.

v3.1.CO2 (03.09.2018)
Algorithms: SCIAMACHY BESD v02.01.02, GOSAT ACOS v7.3.10a, GOSAT NIES v02.72, GOSAT NIES v02BC, GOSAT PPDF-S v02, GOSAT RemoteC v2.3.8, GOSAT UOL-FP v7.2
Minimum number of algorithms used for median calculation: variable, typically number of algorithms minus two.
Remarks: New retrieval versions, algorithm composition now variable in time allowing longer time series, new common a priori (SECM 2018), re-structured code.

v2.2c (04.10.2016)
Algorithms: ACOS v7.3r01, NIES v02.21, RemoteC v2.3.8, UOL-FP v7.0
Minimum number of algorithms used for median calculation: 3 (of 4)
Remarks: Based on v2.0c but with new retrieval versions, without NIES PPDF-S, and new common a priori (SECM 2016).

v2.2b (04.10.2016)
Algorithms: ACOS v7.3r01, BESD v02.01.02, NIES v02.21, RemoteC v2.3.8, UOL-FP v7.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Based on v2.0b but with new retrieval versions, without NIES PPDF-S, and new common a priori (SECM 2016).

v2.2a (04.10.2016)
Algorithms: ACOS v7.3r01, BESD v02.01.02, NIES v02.21, RemoteC v2.3.8, UOL-FP v7.0, WFMD v4.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Based on v2.0a but with new retrieval versions, without NIES PPDF-S, and new common a priori (SECM 2016).

v2.1c (21.01.2016)
Algorithms: ACOS v3.5r02, NIES v02.21, NIES PPDF-S v02.11, RemoteC v2.3.7, UOL-FP v6.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Same as v2.0c but with new retrieval versions.

v2.1b (21.01.2016)
Algorithms: ACOS v3.5r02, BESD v02.01.02, NIES v02.21, NIES PPDF-S v02.11, RemoteC v2.3.7, UOL-FP v6.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Same as v2.0b but with new retrieval versions.

v2.1a (21.01.2016)
Algorithms: ACOS v3.5r02, BESD v02.01.02, NIES v02.21, NIES PPDF-S v02.11, RemoteC v2.3.7, UOL-FP v6.0, WFMD v3.9
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v2.0a but with new retrieval versions.

v2.0c (27.02.2015)
Algorithms: ACOS v3.5r02, NIES v02.21, NIES PPDF-S v02.11, RemoteC v2.3.6, UOL-FP v5.1
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Same as v1.3c but with new retrieval versions and longer time period.

v2.0b (27.02.2015)
Algorithms: ACOS v3.5r02, BESD v02.00.08, NIES v02.21, NIES PPDF-S v02.11, RemoteC v2.3.6, UOL-FP v5.1
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Same as v1.3b but with new retrieval versions and longer time period.

v2.0a (27.02.2015)
Algorithms: ACOS v3.5r02, BESD v02.00.08, NIES v02.21, NIES PPDF-S v02.11, RemoteC v2.3.6, UOL-FP v5.1, WFMD v3.8
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v1.3a but with new retrieval versions and longer time period.

v1.7c (04.02.2015)
Algorithms: ACOS v3.5r02, NIES v02.21 (no bias correction), NIES PPDF-S, RemoteC v2.3.6, UOL-FP v5.1
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Same as v1.3c but with new retrieval versions.

v1.7b (04.02.2015)
Algorithms: ACOS v3.5r02, BESD v02.00.08, NIES v02.21 (no bias correction), NIES PPDF-S, RemoteC v2.3.6, UOL-FP v5.1
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Same as v1.3b but with new retrieval versions.

v1.7a (04.02.2015)
Algorithms: ACOS v3.5r02, BESD v02.00.08, NIES v02.21 (no bias correction), NIES PPDF-S, RemoteC v2.3.6, UOL-FP v5.1, WFMD v3.8
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v1.3a but with new retrieval versions.

v1.6c (08.01.2014)
Algorithms: ACOS v3.4r03, NIES v02.xx, NIES PPDF-S, RemoteC v2.11, UOL-FP v4.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Same as v1.3c but with new retrieval versions.

v1.6b (08.01.2014)
Algorithms: ACOS v3.4r03, BESD v02.00.08, NIES v02.xx, NIES PPDF-S, RemoteC v2.11, UOL-FP v4.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Same as v1.3b but with new retrieval versions.

v1.6a (08.01.2014)
Algorithms: ACOS v3.4r03, BESD v02.00.08, NIES v02.xx, NIES PPDF-S, RemoteC v2.11, UOL-FP v4.0, WFMD v3.6
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v1.3a but with new retrieval versions.

v1.5c (12.06.2013)
Algorithms: ACOS v2.10, NIES v02.xx, NIES PPDF-S, RemoteC v2.1, UOL-FP v4.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Same as v1.3c but with new retrieval versions.

v1.5b (12.06.2013)
Algorithms: ACOS v2.10, BESD v02.00.08, NIES v02.xx, NIES PPDF-S, RemoteC v2.1, UOL-FP v4.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Same as v1.3b but with new retrieval versions.

v1.5a (12.06.2013)
Algorithms: ACOS v2.10, BESD v02.00.08, NIES v02.xx, NIES PPDF-S, RemoteC v2.1, UOL-FP v4.0, WFMD v3.6
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v1.3a but with new retrieval versions.

v1.4c (04.12.2012)
Algorithms: ACOS v2.10, NIES v02.xx, NIES PPDF-S, RemoteC v2.0, UOL-FP v3.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Same as v1.3c but with new retrieval versions.

v1.4b (04.12.2012)
Algorithms: ACOS v2.10, BESD v01.01.02, NIES v02.xx, NIES PPDF-S, RemoteC v2.0, UOL-FP v3.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Same as v1.3b but with new retrieval versions.

v1.4a (04.12.2012)
Algorithms: ACOS v2.10, BESD v01.01.02, NIES v02.xx, NIES PPDF-S, RemoteC v2.0, UOL-FP v3.0, WFMD v2.2bcv7b
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v1.3a but with new retrieval versions.

v1.3c (15.06.2012)
Algorithms: ACOS v2.9, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: Only tiny modifications over Version 1.2c.

v1.3b (15.06.2012)
Algorithms: ACOS v2.9, BESD v01.00.01, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Only tiny modifications over Version 1.2b.

v1.3a (15.06.2012)
Algorithms: ACOS v2.9, BESD v01.00.01, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0, WFMD v2.2bcv7b
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Version used for the EMMA paper. Only tiny modifications over Version 1.2d.

v1.2d (25.05.2012)
Algorithms: ACOS v2.9, BESD v01.00.01, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0, WFMD v2.2bcv7b
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Same as v1.2a but with new reliability post-processing filter for WFMD.

v1.2c (16.05.2012)
Algorithms: ACOS v2.9, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0
Minimum number of algorithms used for median calculation: 3 (of 5)
Remarks: GOSAT algorithms only. Consistent error estimates scaled to match inner-monthly precision.

v1.2b (16.05.2012)
Algorithms: ACOS v2.9, BESD v01.00.01, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0
Minimum number of algorithms used for median calculation: 4 (of 6)
Remarks: Full physics algorithms only. Consistent error estimates scaled to match inner-monthly precision.

v1.2a (16.05.2012)
Algorithms: ACOS v2.9, BESD v01.00.01, NIES v02.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0, WFMD v2.2bcv7
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Consistent error estimates scaled to match inner-monthly precision.

v1.0 (17.04.2012)
Algorithms: ACOS v2.9, BESD v01.00.01, NIES v01.xx, NIES PPDF-DOAS v1.0, RemoteC v1.0, UOL-FP v3.0, WFMD v2.2b
Minimum number of algorithms used for median calculation: 5 (of 7)
Remarks: Initial version.

Data Access

The data access is password-restricted. Please use the following form to obtain username and password.

Email:

Name:

Institute:

Country:

Please tell us in a few words (>3) what you plan with the data:

I agree with the conditions of use:

After registration, please use the following link to access the data archive:
http://www.iup.uni-bremen.de/~mreuter/emma_data/

Till now, the EMMA CO2 data sets cover the period 01/2003-12/2018.

The data-format is NCDF.

Condition of Use

When using EMMA data, you agree ...

... to inform us prior to any publication where EMMA data products are planned to be used,

... to offer us co-authorship for any planned peer-reviewed publication based on EMMA data products (for non peer-reviewed publications it is sufficient if you add an appropriate acknowledgement),

... not to distribute the EMMA data products to any third party (the only exception being colleagues working in your institute, in this case you agree to inform them about the conditions listed here and that they also have to accept these conditions).

Acknowledgements

EMMA is in part funded by ESA/ESRIN (GHG CCI, GHG CCI-II, ADVANSE), EU (MACC-II, MACC, CityZen, C3S), DLR (SADOS), and the state and the University of Bremen as well as participating institues and funding organizations.