AGU’s Data Policy states that all data necessary to understand, evaluate, replicate, and build upon the reported research must be made available and accessible whenever possible. This FAQ page provides additional recommendations and resources. Information on the history and context of the policy (from the 16 September issue of Eos) is available here. Additional tools for practicing reproducible research are here.
Along with many other publishers, AGU has endorsed the Statement of Commitment of the Coalition on Publishing Data in the Earth and Space Sciences (COPDESS) and also the Transparency and Openness Promotion Guidelines (TOP). Data should, to the greatest extent possible, be stored in appropriate domain repositories that are widely recognized and used by the community, follow leading practices for data curation, and can provide additional data services. A directory of data repositories endorsed by AGU is here.
Q: What are the “best practices” regarding the research data, samples, and code associated with my publication and how can I follow them?
Some best practices are:
- Deposit the data in support of your publication in a leading domain repository that handles such data. A directory of leading data repositories in the Earth and space sciences is here.
- If a domain repository is not available for some of all of your data, deposit your data in a general repository such as Zenodo, Dryad, or Figshare. All of these repositories can assign a DOI to deposited data, or use your institution’s archive.
- Data should not be listed as “available from authors.”
- Make sure that the data are available publicly at the time of publication and available to reviewers at submission—if you are unable to upload to a public repository before submission, you may provide access through an embargoed version in a repository or in datasets or tables uploaded with your submission. (Zenodo, Dryad, Figshare, and some domain repositories provide embargoed access.) Questions about this should be sent to AGU journal staff.
- Cite data or code sets used in your study as part of the reference list. Citations should follow the Joint Declaration of Data Citation Principles.
- Develop and deposit software in GitHub and cite that or include simple scripts in a supplement. Code in Github can be archived separately and assigned a DOI through Zenodo for submission.
- Identify your samples with IGSNs.
- List all funding sources (including in-kind support) in the Acknowledgments and indicate grant numbers and funders in your submission in GEMS (we help connect your work to your funders).
- Include ORCIDs for all authors.
Use and follow these resources to develop projects that support reproducibility and integrity in research.
Q: What data must be deposited?
In general, you should deposit and archive the final derived data that you report in your paper. We recognize that there are diverse data types in the Earth and space sciences, and in general the data that are meant to be archived are the data types that most specific repositories accept. These are usually not the raw data collected at instruments in most cases unless repositories are specifically set up to include those data.
For modeling papers, usually the code and key run file information should be archived along with representative outputs that form or illustrate the key results or images in the paper.
Q: What does AGU’s policy require of authors?
The Data Policy requires authors to:
- At submission, include a statement in the Acknowledgments section of the manuscript indicating where readers can find the data that support or underlie the conclusions presented in the manuscript. The data can be in repositories (see above), the references, tables, figures, or supplemental material. “Data available from authors” is not allowed. Failure to include such a statement in the Acknowledgments or stating that data are “from the authors” or “upon request” will cause your submission to be returned to you.
- Confirm data policy compliance in the online GEMS submission form.
- Explain any unavailable data in both the Acknowledgments section of the manuscript AND the Cover Letter (see below question). In general, lack of availability of data will preclude publication in an AGU journal.
- Cite published data sets in the references according to the Joint Declaration of Data Citation Principles. See our citations guidelines for more information. Some examples of data citations are here.
- See also specific examples on how to comply, below.
Q: What are the various ways my submission can comply with AGU’s Data Policy?
Below are ways you can comply with our data policy (multiple ways may apply). In general, if any your data has not been made available using one of the methods below, your submission will be sent back to you. Please see other sections of the FAQ for additional details.
- The authors have deposited the data in a domain or general repository. The References cites the deposited data set(s) and a statement with a link to the dataset is included in the Acknowledgements.
- Data deposition in a domain repository or general repository is underway but incomplete. Justification/explanation is provided in the Cover Letter field of the submission form. The authors have uploaded a temporary copy of the data as file type “Related Manuscript File,” and understand that if the manuscript is accepted data deposition must be completed before publication. Explanation of this is in the Acknowledgements.
- The data are contained in a published source (not a repository) cited in the References, and the Acknowledgments section notes this.
- The data are curated by a third party. The Acknowledgments section clearly describes the restrictions, who controls access to the data, and how reviewers and readers can request a copy from the third party.
- The authors confirm that there is no relevant domain or general repository for the data, and the data are too large to include as Supporting Information. The authors have arranged for the data to be in an institutional repository. The Acknowledgements includes a description of the transparent process for reviewers and readers to access the data.
- The paper is theoretical and no data are used. The Acknowledgements section notes this.
- None of the above – please describe in the cover letter and Acknowledgements. Include in your description information about file types and sizes. Other sections in this FAQ explain how to handle unique cases.
Q: If my data are in a repository, what should I do?
If your data are in a repository, and they have received a DOI or can be cited, please cite them as any other reference. The reference should include an identifier that links directly to the data sets. Include a statement in the Acknowledgments that the data sources are cited in the references.
Q: What if my data are not available at the time of submission but will be at time of publication?
AGU requires that the data be available at the time of publication, and be available in some form for the reviewers to evaluate, such as through an embargoed version in a repository (preferred) or datasets or tables uploaded with your submission. If your data will be placed in a repository later but before publication, please indicate that in the Cover Letter and Acknowledgements section, as well as how reviewers can access it. We will hold publication until the data are deposited. The Acknowledgments section should be updated accordingly before publication.
Q: What if datasets are too large to upload to the submission or to a repository?
If zipped data files are too large to upload to a domain repository, please consult with your institution, or consider if a smaller subset of data needed to reproduce the results can be uploaded. If not, please notify staff in your cover letter or by contacting the journal email address and explain the size limitations and the scope of the data.
Q: What if my study uses model output?
In general, you should deposit the code and run parameters or run files and, where possible, include output that is used in or underlies the representative results (images) in the paper. We encourage posting code on publicly accessible sites such as GitHub, with a static presentation in Zenodo, or other code repositories. We encourage the use of an MIT or similar license for sharing software.
Q: What if my data is owned by a third party, and they do not allow data to be released?
AGU, like other leading publishers, does not allow data to be held secretly by third parties nor data to be available from authors through ad-hoc requests. Much research shows that these models regularly fail to meet data access requirements over time including soon after publication.
If data are provided or owned by a third party and not curated, release of the data (needed to support your results) should be discussed and agreed on with the third party before submission. At submission, your data statement should explain how readers can obtain access to these. If the data were obtained from a third party who provides curation, the readers should be able to equally access the data as the authors did (i.e., preferred access to some, but not all, individuals is not acceptable except for cases where standard exceptions apply and where there are protocols such as privacy); please add to your Acknowledgements the reasons for restricted access and how the reader can access the data,. This would be the case for the purchase of data from a company or institution (e.g., satellite scenes). In cases where data is restricted for privacy reasons, the data statement might read, “Access to the data are restricted by law for privacy purposes as required by our IRB, and readers must apply for access; the data are curated according to the IRB’s guidelines, which are available here [link to guidelines].” Another viable, and preferred option, would be that the third party publish the data themselves (in a repository or separate data paper); the manuscript can then cite that data.
Q: What if data are taken from references?
Simply state in the Acknowledgments that the data used are from cited references. Please list all references in the main reference list (references do not count toward length limits).
Q: What if my paper is theoretical and/or contains no new data or modeling?
If your paper contains no new data, simply state that there are no new data or models in the acknowledgments.
Q: Under which circumstances do I need to provide information about data in the cover letter?
Explanations of data availability must be added to the Cover Letter when you are unable to provide access to your data at the time of submission and/or by publication. In general, lack of availability of data will preclude publication in an AGU journal. See other sections of this FAQ for specific circumstances.
AGU reserves the right to refuse publication when authors are unwilling to make the underlying data available or otherwise refuse to comply with the letter and spirit of the policy. If you do not provide a data access statement, your submission may be declined. This decision resides with the editor-in-chief of the journal.