Skip to main content
U.S. FlagAn official website of the United States government
icon dot gov
Official websites use .gov

A .gov website belongs to an official government organization in the United States.

icon https
Secure .gov websites use HTTPS

A lock () or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

OCO-2 v11 MIP
Flux Maps Download

The OCO-2 v11 MIP


Input files

  • Protocol document (version 8.3.1; latest as of 12 Mar 2025)
  • Observational constraints

    ObsPack dataobspack_co2_1_OCO2MIP_v5.0.1_2024-11-14
    OCO-2 dataOCO2_b11.2_10sec_GOOD_r2.nc4
    OCO-3 dataOCO3_b11_10sec_GOOD_r2.nc4
    TCCON datatccon_ggg2020_timeaverages_20142024.tar.gz

  • Input land, ocean, and fossil fluxes

    • Fossil fuel CO2 emissions
      Download: https://zenodo.org/records/13947942 - be sure to use version 2024.3

      In order to prevent differences due to the choice of fossil fuel burning emissions, all modeling groups participating in the OCO-2 v11 MIP must use the same of FF emissions for all experiments in the MIP. All groups must use these FF fluxes for 2014 through 2024. We suggest using them for as many years as possible before 2014 when generating the prior 3D CO2 mixing ratio field used at the start time of the inversion runs. FF fluxes going back to the year 2000 are available from this Zenodo site. The FF emissions are in the form of daily netCDF files, with 24 hourly fluxes (in terms of GMT) on a 1 degree x 1 degree lat/lon grid. The emissions are to be used as given inside of each hour, with no interpolation performed in time. Each group will map the 1x1 degree fluxes to the horizontal grid of their own transport model. They should ensure that the global total emission for each hour matches that given in the md5sums.txt file provided. In the FF emissions files, variable 'hourly_total_emission' gives the flux inside each grid box in units of [TgC/hour]; the area for each box is given as a convenience, to help convert these to flux densities in units of [TgC/m^2/hour] for use in mapping the fluxes to the users' horizontal grids. Each user should check that their mapping preserves the integral mass flux, at least at the global scale. File carbonmonitor-global_datas_2024-10-15.nc may be used to check the mass balance at the scale of 13 large countries or groups of countries.

    • Air-sea CO2 exchange
      Download: Median_best7GCP_ocean_dataproducts_9024.nc

      • This is the median of the seven models that were used by the Global Carbon Project (2023) to calculate the pCO2-based air-sea flux estimates on their spreadsheet for 1990-2002: CMEMS-LSCE-FFNN, Jena-MLS, JMA-MLR, LDEO-HPD, NIES-ML3, SODA-ETHZ, and SOM-FFN. (Note that we have excluded the UoEx-Watson model, as the GCP did.). If some of the models did not provide flux estimates for certain 1x1 degree grid boxes (e.g., near coastlines), we have taken the median of just those models that did.

      Again, because these come from the GCP data products rather than the ocean models, they tend to have something like a half PgC/yr more ocean uptake than the models do (even after accounting for the river flux difference between the two). Gets up to as high as -2.8 PgC/yr in 2019.

    • MiCASA land and fire emissions
      • Please download the MiCASA land biospheric CO2 fluxes from https://portal.nccs.nasa.gov/datashare/gmao/geos_carb/MiCASA/v1/.

      MiCASA land biospheric CO2 fluxes are available at 0.1° x 0.1° and at 0.5° x 0.5° spatial resolution at three temporal resolutions:

      1. 3-hourly: NEE and NPP only
      2. Daily: NPP Rh ATMC NEE FIRE FUEL
      3. Monthly: NPP Rh ATMC NEE FIRE FUEL
      Positive NPP indicates uptake by vegetation. Positive Rh indicates emission to the atmosphere. NEE = Rh - NPP - ATMC, and NBE = NEE + FIRE + FUEL. ATMC adjusts net exchange to account for missing processes and better match long-term atmospheric budgets.

      NOTE: 3-hourly Rh may be computed from 3-hourly NEE and NPP as Rh=NEE+NPP plus daily ATMC

      The OCO v11 MIP will be running some of these tracers FORWARD through the models, to test transport differences, and for use as prior land biospheric fluxes in a common-prior inversion branch study.

      Brad Weir announcement of 9/25/2024:

      Dear All,

      We’re happy to announce that MiCASA v1 is now available at https://portal.nccs.nasa.gov/datashare/gmao/geos_carb/MiCASA/v1

      It will be available on GES DISC soon as well.

      MiCASA is an extensive upgrade of CASA-GFED3 diagnostic biospheric flux estimates (NPP, Rh, FIRE, and burning of harvested coarse woody debris) produced at NASA GSFC and used as a prior in many OCO MIP submissions. MiCASA runs at a native daily timestep and 0.1 degree horizontal resolution. It covers 2001-2023 with estimates for the first half of 2024 and a near-real-time version coming soon. This release also includes an estimate of the correction needed to match atmospheric budgets (ATMC). It is recommended this be used as a correction to Rh and included in NEE calculations, i.e., NEE = Rh - ATMC - NPP, but is included as a separate variable so the user can choose.

      More information is available at https://portal.nccs.nasa.gov/datashare/gmao/geos_carb/MiCASA/v1/MiCASA_README.pdf.

      Best, Brad

      P.S. For the beta users out there, the only change is that variables are now in units of kg C m-2 s-1 instead of g C m-2 day-1.

    • SiB4 land and fire emissions

      • The SiB4 land biospheric fluxes at 0.5° x 0.5° spatial and hourly temporal resolution are available in daily files at https://www.gml.noaa.gov/aftp/user/kaushik/SiB4_priors_for_OCO-MIP.

      Variables are assim (=GPP), resp_auto, resp_het, resp_tot, resp_fire. Users can calculate NEE, NPP, or NBP as needed from these.

      Flux units are micromol CO2 m-2 s-1.

      Files should be downloadable with wget. You can use the syntax:

      wget -i sib4_wget_filelist.txt

      The list of files to download is sib4_wget_filelist.txt.

SF6 MIP
SF6 Model Intercomparison Project protocol


Older Input files