Welcome to the Community Inversion Framework!#
The Community Inversion Framework (CIF) is an open-source Python package for estimating greenhouse-gas (GHG) and air-pollutants surface fluxes from atmospheric observations. It connects multiple atmospheric transport models with several inversion algorithms — variational 4D-Var, analytical, and ensemble methods — through a unified plugin system, enabling reproducible and interoperable inversion studies across the atmospheric science community.
Recent developments also include inter-disciplinary studies coupling atmosphere and land-process models (e.g., coupling with ORCHIDAS, or for CH4-DAS).
The development of CIF was initiated in 2018 with initial support from the European Union H2020 project VERIFY. It has since been supported by several projects including CoCO2, CHE, and ARGONAUT. It is currently supported by EYE-CLIMA and the Copernicus Atmospheric Monitoring Service.
License
The Community Inversion Framework is governed by the CeCILL-C license under French law. Please consult the reference text here for further detail. The license grants full rights for the users to use, modify and redistribute the original version of the CIF, conditional to the obligation to make their modifications available to the community and to properly acknowledge the original authors of the code.
Use, acknowledgement and citation
The Community Inversion Framework is the result of a collective effort by a community of scientists who have chosen to openly share their developments. Continued maintenance and new features are made possible by the core development team and their funding agencies.
Any scientific use of CIF must be acknowledged. At minimum, please include the following sentence in the acknowledgements section of your publication:
“This study used the Community Inversion Framework (CIF; Berchet et al., 2021, https://doi.org/10.5194/gmd-14-5331-2021), an open-source Python package for atmospheric inversion studies.”
If CIF contributed substantially to the scientific results — through specific developments, significant technical support, or novel methodological components — the authors are requested to contact the development team to discuss co-authorship and additional citations (see Publications).
Install PyCIF and its system dependencies on your platform, or pull the ready-to-run Docker image.
Run your first forward simulation and inversion in a few steps.
Learn how CIF is structured: the plugin system, inversion workflow, and key concepts.
Full documentation of all plugins, their parameters, and how to wire them together in a YAML file.
Step-by-step guides covering forward runs, variational inversions, adjoint tests, and ensemble methods.
Ready-to-use configuration files for common use cases.
Reference papers and publications by CIF developers and users.
How to report issues, add a new plugin, or contribute to the documentation.
In-depth guides for implementing new models, transforms, and inversion methods.
Help Desk#
We do our best to make this website as comprehensive as possible.
However, should you find no answer to your questions, please contact: help@community-inversion.eu
It is also possible to register to the information mailing list dedicated to CIF:
info-cif@lists.lsce.ipsl.fr.
Important information, as well as general discussions, are posted on that list.
Only subscribers can post and receive corresponding mails.
To subscribe, send an empty email to
sympa@lists.lsce.ipsl.fr
with the following subject line: SUBSCRIBE info-cif@lists.lsce.ipsl.fr