Atmospheric inversions
General description: Atmospheric CO2 inversions estimate surface-to-atmosphere net carbon fluxes by utilizing atmospheric CO2 concentration measurements and an atmospheric transport model (see more details). The contributing inversions are recent state of the art systems that have been used for the RECCAP synthesis and compared in Peylin et al., Biogeosciences (2013). The inversion results have been provided by individual scientists, as credited, who have made them available and encourage their use.
The original data have been reformatted onto a 1x1 degree grid and further processed to correct for differences in fossil fuel emissions between inversions. Only the natural land and ocean fluxes are displayed under this portal. If you plan to use any of these results, you need to contact each modeling group. Note that regional totals and reformatted 3-D fluxes can be accessed under the “Transcom” web site (http://transcom.lsce.ipsl.fr/) after contacting Philippe Peylin (Peylin@lsce.ipsl.fr).
The use of data is conditional on citing the original data sources. Full details on how to cite the data are given for each inversion in the references below and in the corresponding web links. The Global Carbon Project facilitates access to data to encourage its use and promote a good understanding of the carbon cycle. Respecting original data sources is key to help secure the support to enhance, maintain and update valuable data.
MACC-II:
Contacts: Frederic Chevallier (frederic.chevallier@lsce.ipsl.fr)
Description: MACC-II corresponds to version 11.2 of the CO2 inversion product from the Monitoring Atmospheric Composition and Climate - Interim Implementation (MACC-II) service (http://www.gmes-atmosphere.eu/). It is based on a variational formulation. Fluxes are solved at the resolution of the transport model (LMDz-v4 : 2.5° x 3.75°) and a 8-day daytime/nighttime resolution. Prior land and ocean fluxes come from the ORCHIDEE land surface model climatology and Takahashi et al. (2009), respectively. Raw atmospheric CO2 concentrations from 134 observing sites are used.
References:
Chevallier, F., et al.: CO2 surface fluxes at grid point scale estimated from a global 21 year reanalysis of atmospheric measurements, J. Geophys. Res., 115, D21307, doi:10.1029/2010JD013887, 2010.
Data access: http://www.gmes-atmosphere.eu/d/services/gac/delayed/
JENA-MPI:
Contacts: Christian Roedenbeck (Christian.Roedenbeck@bgc-jena.mpg.de)
Description: The Jena inversion has been designed to estimate interannual variations of land and ocean fluxes, based on raw CO2 concentration data from 50 observing sites. It uses a variational approach with the TM3 transport model (4° x 5°, using interannually-varying winds). Fluxes are optimized at the spatial resolution of the transport model and at roughly weekly to interannual time scales. Prior land fluxes come from fossil fuel emissions and a modelled mean biosphere pattern, and ocean fluxes from an inverse estimate based on ocean carbon data (Mikaloff-Fletcher et al. 2006).
References:
C. Rödenbeck et al.: CO2 flux history 1982-2001 inferred from atmospheric data using a global inversion of atmospheric transport, Atmos. Chem. Phys. 3, 1919-1964 (2003) ; Updated version s96_v3.51
Data access: Results can be downloaded from “http://www.bgc-jena.mpg.de/~christian.roedenbeck/download-CO2/”
LSCE-ana:
Contacts: Philippe Peylin (Peylin@lsce.ipsl.fr)
Description: LSCE-ana inversion is based on a Bayesian synthesis method described in Peylin et al. (2005) and corresponds to the results described in Piao et al., 2009. Fluxes are optimized at the resolution of the transport model (LMDz-v3 : 2.5° x 3.75°) on a monthly base. Prior land and ocean fluxes come from ORCHIDEE land surface model climatology and Takahashi et al. (2002), respectively. Monthly mean CO2 concentrations from 74 observing sites are used.
References:
Peylin, P., et al.: Daily CO2 flux estimates over Europe from continuous atmospheric mea- surements: 1, inverse methodology, Atmos. Chem. Phys., 5, 3173–3186, doi:10.5194/acp-5-3173-2005, 2005.
Piao, S. L., Fang, J. Y., Ciais, P. Peylin, P., Huang, Y., Sitch, S., and Wang, T.: The carbon balance of terrestrial ecosystems in China, Nature, 458, 1009–1014, 2009.
Data access: contact Philippe Peylin
CCAM:
Contacts: Rachel Law (rachel.law@csiro.au)
Description: CCAM inversion uses a Bayesian synthesis method, described in Rayner et al. 2008. It is based on CCAM transport model but with not interannual-varying winds. Fluxes are solved for 94-land and 52-ocean regions on a monthly base. Prior land and ocean fluxes come from CASA land surface model climatology and Takahashi et al. (1999), respectively. Monthly mean observations from 73 atmospheric CO2 records and 7 d13C records are used.
References:
Rayner, P. J., Law, et al.: Interannual variability of the global carbon cycle (1992–2005) inferred by inversion of atmospheric CO2 and d 13 CO2 measurements, Global Biogeochem. Cy., 22, GB3008, doi:10.1029/2007GB003068, 2008.
Data access: contact Rachel Law
MATCH:
Contacts: Peter Rayner (prayner@unimelb.edu.au)
Description: MATCH inversion uses a Bayesian synthesis method, described in Rayner et al. 2008. It is based on MATCH transport model but with not interannual-varying winds. Fluxes are optimized for 67-land and 49-ocean regions on a monthly base. Prior land and ocean fluxes come from CASA land surface model climatology and Takahashi et al. (1999), respectively. Monthly mean observations from 73 atmospheric CO2 records and 7 d13C records are used.
References:
Rayner, P. J., Law, et al.: Interannual variability of the global carbon cycle (1992–2005) inferred by inversion of atmospheric CO2 and d 13 CO2 measurements, Global Biogeochem. Cy., 22, GB3008, doi:10.1029/2007GB003068, 2008.
Data access: contact Peter Rayner
CTE2013:
Contacts: Wouter Peters (wouter.peters@wur.nl)
Description: CTE2013 inversion corresponds to the CarbonTracker system, described in Peters et al. (2007) with a specific zoom over Europe and North America. It is based on TM5 transport model with two nested grids (1°x1° resolution over Europe/North America and 3° x 2° elsewhere) and with interannually-varying winds. It uses an ensemble Kalman filter to optimized weekly fluxes over 138-land and 30-ocean regions. Prior land and ocean fluxes come CASA biosphere model (including interannual flux variations) and from the ocean inversions reported in Jacobson et al. (2007), respectively. Raw CO2 concentrations from 117 observing sites are used.
References:
Peters et al. An atmospheric perspective on North American carbon dioxide ex- change: CarbonTracker, P. Natl. Acad. Sci. USA, 104, 18925– 18930, doi:10.1073/pnas.0708986104, 2007.
Jacobson, A. R., et al.: A joint atmosphere-ocean inversion for surface fluxes of carbon dioxide: I. Methods and global-scale fluxes, Global Biogeochem. Cy., 21, GB1019, doi:10.1029/2005GB002556, 2007.
Data access: contact Wouter Peters and access via http://www.carbontracker.eu/.
RIGC:
Contacts: Prabir Patra (prabir@jamstec.go.jp)
Description: RIGC inversion results are based on the TransCom Level 3 Bayesian synthesis method, using the NEIS/FRCGC transport model with interannually-varying winds. They are described in details in Patra et al. (2005). Fluxes are optimized for 42-land and 22-ocean regions on a monthly base. Prior land and ocean fluxes come from CASA biosphere model (climatology) and Takahashi et al. (1999), respectively. Monthly mean CO2 concentrations from 74 observing sites are used.
References:
Patra, P. K., et al.: Role of biomass burning and climate anomalies for land-atmosphere carbon fluxes based on inverse modeling of atmospheric CO2, Global Biogeochem. Cy., 19, GB3005, doi:10.1029/2004GB002258, 2005.
Data access: contact Prabir Patra
TRCOM-mean:
Contacts: Kevin Gurney (kevin.gurney@asu.edu)
Description: The TRCOM-mean results are based on the TransCom 3 Level 2 analysis found in Gurney et al. (2008), based on a Bayesian synthesis method. The individual posterior flux results from 11 transport models (using no interannually-varying winds) are averaged to generate the multi-model mean. Fluxes are optimized for 11-land and 11-ocean regions (sub-continental size) on a monthly base. Prior land and ocean fluxes come from CASA biosphere model (climatology) and Takahashi et al. (1999), respectively. Monthly mean CO2 concentrations from 103 observing sites are used.
References:
Gurney, K. R., et al., Interannual variations in continental-scale net carbon exchange and sensitivity to observing networks estimated from atmospheric CO2 in- versions for the period 1980 to 2005, Global Biogeochem. Cy., 22, GB3025, doi:10.1029/2007GB003082, 2008.
Data access: contact Kevin Gurney
NICAM-TM:
Contacts: Yosuke Niwa (yniwa@mri-jma.go.jp)
Description: NICAM inversion is based upon the TransCom 3 Bayesian synthesis method, further described in Niwa et al. (2012). It relies on the NICAM-TM transport model with interannually-varying winds. Fluxes are optimized for 29-land and 11-ocean regions on a monthly base. Prior land and ocean fluxes are taken from the CASA biosphere model (climatology) and Takahashi et al. (2009), respectively. Monthly mean CO2 concentrations from 59 surface sites and from 12 aircraft points of CONTRAIL are used.
References:
Niwa, Y., et al.: Imposing strong constraints on tropical terrestrial CO2 fluxes using passenger aircraft based measurements, J. Geophys. Res., 117, D11303, doi:10.1029/2012JD017474, 2012.
Data access: contact Yosuke Niwa
JMA:
Contacts: Takashi Maki (tmaki@mri-jma.go.jp)
Description: JMA inversion is based upon the TransCom 3 Bayesian synthesis method, further described in Maki et al. (2010). It relies on the JMA-CDTM transport model with interannually-varying winds. Fluxes are optimized for 11-land and 11-ocean regions on a monthly base. Prior land and ocean fluxes are taken from the CASA biosphere model (climatology) and Takahashi et al. (2002), respectively. Monthly mean CO2 concentrations from 146 sites are used.
References:
Maki et al. New technique to analyse global distributions of CO2 concentra- tions and fluxes from non-processed observational data, Tellus B, 62, 797–809, doi: 10.1111/j.1600-0889.2010.00488.x, 2010.
Data access: contact Takashi Maki
Disclaimer:
This material is general in nature. It is made available on the understanding that the Global Carbon Project (GCP) is not thereby engaged in rendering professional advice. Before relying on the material in any important matter, users should carefully evaluate its accuracy, currency, completeness and relevance for their purposes, and should obtain any appropriate professional advice relevant to their particular circumstances.
In some cases the material may incorporate or summarise views, guidelines or recommendations of third parties. Such material is assembled in good faith, but does not necessarily reflect the considered views of the Global Carbon Project (GCP), or indicate a commitment to a particular course of action. Links to other websites are inserted for convenience and do not constitute endorsement of material at those sites, or any associated organisation, product or service.