Authorship of NutNet Manuscripts

NutNet is a sociological, as well as an ecological, experiment

NutNet exists because many scientists from around the world cooperate to generate data to make global ecological analysis and discovery possible. By joining NutNet, you are not just gaining access to a long-term, global dataset, you are also gaining a global network of collaborators. Because participation in NutNet includes both high quality data and high quality collaborators, writing scientific papers necessitates deeper consideration of the authorship process than is needed when working with a database or more traditional data collection projects. Here we lay out ground rules that establish a fair, and well-tested, process for determining authorship. We aim to be inclusive while not diluting the value of authorship on a manuscript. Much of our philosophy and approach to authorship in NutNet is also published in Borer et al. 2023. Please engage with the NutNet manuscript writing process knowing you are helping to forge a new model of collaborative ecology.

Who can be an author on a NutNet Manuscript?

Anyone who follows these guidelines can participate in NutNet and write a manuscript using NutNet data. You do not need to lead a site, contribute data, or know a NutNet PI to join the Network, gain access to NutNet data, and write a manuscript that leverages the effort of the NutNet scientists and the data we are producing.

Graduate students and postdocs are encouraged to author NutNet papers. Early career scientists have played myriad roles in leading NutNet research, including: leading papers that focus on unique response variables they have measured at a single site or a cluster of regional sites; leading network-level papers; or leading papers arising from add-on studies (collecting samples to quantify new response variables that answer research questions that go beyond the core NutNet data). We welcome graduate students and other early-career researchers to contribute in a way that is meaningful for them. Students new to the network or those seeking extra advice are welcome to contact the early-career researcher liaison within the NutNet authorship committee, Lauren Sullivan (LLSULL@msu.edu).

NutNet data types: two paths to a NutNet publication

As you consider how to engage with the data being produced by NutNet scientists, it is important to know that NutNet has two major data types: published data and unpublished (not yet published) data.

1. New analyses of published NutNet data. The NutNet experimental data and infrastructure have been used to generate well over a hundred publications, to date. NutNet data have been published along with papers and are openly accessible. These data are a valuable resource for the scientific community to tackle new questions.

By conceiving a new idea that can be tested using published NutNet data, this path requires only that you cite the doi of the published dataset(s) you use. There are no additional co-authorship expectations.

This option is recommended for synthesis working groups (e.g., sDiv, NCEAS, LTER).
Graduate students, or anyone participating in NutNet, but on a relatively short timeline, also can benefit from asking new questions using published project datasets because this approach does not involve additional coauthors so can proceed more quickly than using unpublished project data.
While project knowledge (i.e., working with a NutNet scientist) could help inform analyses and avoid misuse or misinterpretation of project data, you are not required to work with the NutNet collaborator group. You may publish without including NutNet participants as authors. We do request that you send us a link to any published paper that includes NutNet data.

2. Analyses of unpublished NutNet data. NutNet is an ongoing global experiment, so new data are constantly being collected, submitted, quality checked, and uploaded into the collaborative dataset. By conceiving a new idea that requires project data that have not yet been published, this path may lead to greater novelty, but it also comes with additional expectations, namely to provide opportunities for the scientists producing data to opt in as coauthors.

Because of the distributed design of this experiment, whenever an analysis can include data from all (or most) sites in the network, it should do so; sub-network and site-level analyses should be reserved only for questions that can not be answered using data from all contributing sites.

The rest of this document focuses on philosophy, expectations, and mechanics for publications arising from analysis of unpublished NutNet data.

Earning authorship through engagement: what is a substantial contribution?

In NutNet, authorship must be earned by making at least two types of substantial contributions to each manuscript.

Documenting contributions: the authorship rubric

Contribution areas of each individual to each manuscript are documented in the form of an authorship rubric that is published with each paper. Authorship-worthy contributions in NutNet include traditionally recognized intellectual contributions (e.g., conceived idea, performed analyses, wrote manuscript). However, the NutNet collaboration also involves time and financial contributions of scientists around the world without whom the data at the core of the project would not exist. Therefore, NutNet collaborators have agreed that actively leading a site from which data are being used in a paper also represents a substantial contribution.

Site leadership - who is a site level coordinator?

Serving as a current lead of a NutNet site whose data are used in a manuscript is considered a substantial contribution. A site lead (or two), sometimes referred to as ‘site PI’ or ‘site level coordinator,’ manages site permissions or fees and ensures data collection, proofing, and submission. Because one goal of NutNet is to foster an engaged community of scholars making ongoing contributions to the project, once an individual steps out of this role, they are no longer considered a site coordinator for authorship purposes. However, they may continue to ‘check the box’ (see rubric, below) as site level coordinator for up to three years after stepping out of their site-level coordinator role to acknowledge their prior contributions of e.g., time and financial investment. After three years, they cannot check this box and must contribute to a manuscript in at least two other ways. Former site coordinators interested in contributing intellectually to a manuscript, but are unsure how, should contact the lead author for guidance.

We note here that we separate manuscript authorship from data authorship in NutNet (see Overview of NutNet authorship, Step 9, below). By separating data authorship from manuscript authorship, we give credit for the time, effort, and expense of maintaining a site and contributing the data used in a NutNet publication, but we separate this from the additional intellectual contributions involved in the co-authorship of individual manuscripts. Any current or former site coordinator will be listed as an author of the dataset, even after they can no longer claim this role for opting into the authorship of manuscripts.

Other types of authorship-worthy contributions include initiating and framing the project or question, analyzing the data, checking code, or writing or editing the manuscript (see below for a full list of authorship-worthy contributions). Collaborators who are not serving as the coordinator of a site with data used in a paper can qualify as an author on a paper by contributing to the manuscript in two other substantial ways, for example, by (1) developing the ideas behind a paper and (2) analyzing the data. Many individuals who have not led a site or generated any NutNet data have earned authorship by leading or contributing in meaningful ways to NutNet publications.

The NutNet Authorship Rubric

Lead authors are responsible for ensuring that when NutNet manuscripts are published, they include an authorship rubric table that indicates authorship contributions.You can find an Authorship Contribution Template at this link. For opt-in* papers, every co-author is expected to have at least two of the following areas checked in the authorship rubric.

Rubric item	Example contribution meriting a checked box
Developed and framed research question(s)	Originated idea for current analysis of Nutrient Network data; contributed significantly to framing the ideas in this analysis at early stage of manuscript development
Analyzed data or generated new dataset	Generated models (conceptual, statistical and/or mathematical), figures, tables, maps, etc.; generated a Network-scale dataset being used in this manuscript’s analysis (e.g. soil chemistry, root biomass, etc.)
Contributed to data analyses	Provided new code or specific suggestions for data analysis; checked code for core analyses
Wrote the paper	Wrote the majority of at least one section of the paper
Contributed to paper writing	In addition to editing text for clarity, also provided suggestions such as restructuring ideas or logical flow, or text and citations linking to new literature areas
Current site- level coordinator	Populate this box from the central NutNet database. Currently coordinating and funding site permissions, experimental maintenance, data collection, proofing, and submission of data for at least one site providing data used in this manuscript (a.k.a. “Site PI”); former site coordinators should not check this box.
Current network- level coordinator	Populate this box from the central NutNet database. Ongoing substantial contributions (i.e., more than 300 hours per year) to network level activities such as management of the network database, recruiting and assisting sites, network-wide strategic planning and communication, acquiring funding for network level management

*opt-in vs opt-out publications in NutNet:

At the outset of the NutNet project (2006-2007), we laid out three core questions arising from the three project datasets that became the heart of three initial publications. In recognition that the generation and provision of data by scientists around the world was the foundation of this project, we considered data contribution alone as sufficient for co-authorship on these three initial ‘opt-out’ papers.

However, with the philosophy that NutNet is fostering an engaged community of scholars making ongoing contributions to the project – paired with the desire to retain the role of an author as a meaningful contribution – all subsequent papers arising from project data have required a minimum of two areas of contribution to each publication. Authors must ‘opt-in’ by engaging intellectually and investing time on each paper on which they want to be listed as a co-author.

Overview of the NutNet authorship process

The philosophy and mechanics of authorship in NutNet are laid out in Borer et al. 2023. Please read this paper, if you are planning on leading or contributing to a NutNet paper.

The primary goals of the NutNet authorship process are to

minimize overlap and maximize novelty of analyses using unpublished NutNet data;
consistently, accurately and transparently attribute the contribution of each author on each paper;
encourage participation in manuscripts by interested scientists;
ensure that each author has made sufficient contribution to each paper to warrant authorship.

Step 1: Carefully read these guidelines

Steps are also laid out in the flowchart below.

Step 2: Idea generation and suitability of NutNet data for your question

Assess whether NutNet data are appropriate for addressing your scientific question by exploring the metadata and data (if you are already affiliated with NutNet and have access to our shared data). Contact the Network Coordinator or a member of the Authorship Committee (info below) with any questions.

If you are part of a synthesis working group, we recommend that you use only published NutNet data, which means you can stop reading here. Please carefully read the section “NutNet data types: two paths to a NutNet publication” above.

If you are not a scientist contributing to the Nutrient Network dataset, but want to participate in writing a paper with the full group and full dataset, you are welcome to propose an idea. Please reach out to a member of the Authorship Committee after you have an idea in mind and you have decided the data are right for you.

Step 3: Check the abstract database and published papers to determine novelty

For those working with NutNet data that have not yet been published, the next step is to check the abstract database to see who is working on related topics. If you find abstracts with similar concepts or analyses, we recommend you reach out to the lead authors on those abstracts to assess the novelty of your ideas and consider whether the ideas are distinct or if it would be better to combine efforts.

Step 4: Write and submit a project abstract

Submit an abstract. Before you embark on in-depth analyses, we require you to submit an abstract for review by the NutNet Authorship Committee. This abstract should include a clear articulation of your research question, the data/analyses you will use, and a novelty statement describing in detail how your proposed project differs from published and ongoing work within NutNet. Abstracts should be submitted at this link - if you do not have access to submit an abstract, contact the Network Coordinator.

The novelty statement is an important part of this step. Carefully articulate the conceptual novelty that your work provides, distinguishing it from other in-progress and published works using NutNet data. To do this, you should look at both the NutNet abstract database and published papers. This is important for the internal review process and also a useful statement to include in publications to help readers distinguish among the many NutNet papers.

The Authorship Committee. The Authorship Committee keeps track of every proposed, ongoing, and published network paper (as best we can) to minimize overlap, maximize new discovery, and ensure communication among scientists working on related analyses. This committee is not gate-keeping ideas – we are working to maximize collaboration and discovery while avoiding conceptual overlap or scramble competition in this group of >>300 creative scientists.

After reviewing the abstract and novelty statements, the Authorship Committee will work with lead authors to resolve any potential overlap between your ideas and those of other existing projects within NutNet. They will also connect you with others working on similar projects. If and when your abstract is approved by the committee, you will be given access to the NutNet shared data folder, if needed. Return to the website to update the status of your abstract at each step as you progress through this process.

Step 5: Circulate your ideas to allow collaborators to opt in and contribute meaningfully

Only after receiving written approval from the Authorship Committee is it ok to proceed to this step.

→ If you have made it this far and have not yet read Borer et al 2023, please do so now.

The first ‘opt-in’ email is intended to inform the entire network of a paper that is moving forward and to identify who would like to contribute intellectually.

Who receives the opt-in email? For almost all manuscripts, the lead author should circulate the opt-in email to the NutNet listserv <nutnet [at] googlegroups.com>. This allows all participants, even those not interested in a topic or ineligible for co-authorship, to stay informed about NutNet projects. A small subset of manuscripts or analyses will be most appropriate at the single-site or sub-network scale. For these papers (e.g., papers analyzing data from a sub-network “add-on” study or dissertation work with new data collected from one or a few sites), the lead author may choose to circulate the manuscript to only those who contributed data.

The opt-in email should contain the following information (see example here):

The concept as a storyboard, including:

An abstract and outline of the framing, including questions/hypotheses, the general approach, and an outline of the introduction to illustrate concepts and linkages. Provide just enough detail that potential coauthors can see where the paper fits intellectually and how it advances NutNet science.

Figures and tables to illustrate the main results. These are intended to provide a thumbnail sketch of the main predictor and response variables.

A list of site codes, site names, and current site-level coordinators for all sites with data used in the paper’s analysis. If you later add new sites to your analysis after sending the initial opt-in email, contact the leaders of the site(s) that you add, to allow them the chance to opt in.
A reading assignment of 1-2 foundational publications that serve as the intellectual basis for your paper or are otherwise critical for all contributors to read and understand.
A manuscript-related assignment. This is an important first opportunity for coauthors to engage and provide intellectual contributions. The lead author should communicate a clear and direct request that draws on the knowledge and expertise of the coauthors and will help the lead author.

Examples include asking coauthors to propose alternative hypotheses for an observed pattern, describe links to other relevant bodies of literature and add citations, or provide ideas for complementary or alternative statistical analyses.

If there are confusing results or technical issues that make data difficult to interpret, an assignment could be asking coauthors to consider and respond to these. Logical fallacies or problematic statistics are better discovered at this stage than after the weeks or months it may take to write a complete draft.

A link to a shared online spreadsheet with the authorship rubric. Those signing up to contribute to the paper add their contact information and check the areas in which they intend to contribute. A final version of this spreadsheet will be published as supplementary material detailing each author’s personal contribution, for internal and external transparency of contributions (Sauermann and Haeussler 2017).
A deadline for signing up and providing the first contribution. Two weeks (not less) is generally sufficient.

Step 6: Develop the manuscript with co-authors who have opted in

After collaborators indicate their intention to contribute to a paper (“opt-in”), subsequent emails are generally sent only to the subset who have opted in. The goal of each contact should be to provide effective communication with, and solicit specific feedback from, a large group of busy coauthors. Generally, 2-3 whole-group emails are sent by the lead author through the development of a paper. Standard emails during the paper development include:

A summary of input and changes made in response to previous coauthor feedback. Because this process involves the lead author considering many, sometimes conflicting, perspectives, summarizing coauthor responses to each assignment and changes made is a critical part of ensuring that coauthors feel their time was valued, even when specific suggestions were not followed. A point-by-point response is not necessary (nor is it efficient for anyone), but a short description of key changes or how conflicting suggestions were resolved is helpful for everyone. While this process can feel onerous given many suggestions, it serves to improve the work – by the time the final paper is submitted, it has already been critically peer-assessed by many readers.

A new, targeted assignment and deadline. Assignments at each step depend on the needs of the lead author, but generally follow a similar recipe that includes sharing only the information necessary for coauthors to complete their assigned tasks by clearly soliciting key conceptual feedback. For example:

A first contact to the co-authors may include a summary of the first round of coauthor feedback and changes made, plus a draft abstract, introduction, figures, and tables with brief text describing key results. This would then be paired with an assignment and reasonable deadline.

Example assignments include: provide thoughts on points to highlight in the discussion, provide key papers to cite, and recommend a journal to which to submit the paper.

A second contact may include a draft manuscript and instructions to read but not text-edit, to suggest a title, and assess whether the key hypotheses and figures tell the story well.

A third contact may request reviewer suggestions, assistance with formatting, editing, and creating the author and affiliation list (which is a large task for papers with dozens of authors, some with multiple affiliations).

Receiving coauthor edits to the text of a manuscript: We highly suggest sharing the manuscript as a .pdf file and requesting feedback in a shared online spreadsheet with columns for coauthor name, line number(s), and comments. Comments can then be sorted by line number for efficient use by the lead author.

All coauthors must approve of the final version of the manuscript before it is submitted for publication.

chart

Step 7. Confirm author contributions

Before publication, the lead author should carefully review the authorship contribution table to ensure that all authors have contributed at a level that warrants authorship and that contributions are consistently attributed among authors. Has each author made contributions in at least two areas in the authorship rubric? Did each author provide thoughtful, detailed feedback on the manuscript? Authors are encouraged to contact the Network Coordinators or Authorship Committee about any confusion or conflicts.

Step 8. Submit the manuscript

Please ensure the following when submitting your manuscript for publication.

Keywords include “Nutrient Network (NutNet)"
The acknowledgments section contains the following standard text:

This work was generated using data from the Nutrient Network (http://www.nutnet.org) experiment, funded at the site-scale by individual researchers. Coordination and data management have been supported by funding to E. Borer and E. Seabloom from the National Science Foundation Research Coordination Network (NSF-DEB-1042132) and Long Term Ecological Research (NSF-DEB-1234162 and NSF-DEB-1831944) programs, and the Institute on the Environment (DG-0001-13). We also thank the University of Minnesota College of Biological Sciences and the Minnesota Supercomputer Institute for hosting project data and the Institute on the Environment for hosting Network meetings.

Authorship contributions table is included as supplementary material
Borer et al. 2014 is cited as standard methods, and your paper’s methods section is consistent with this paper

Step 9. Publishing NutNet data

Repository: When publishing NutNet data along with a NutNet manuscript, we suggest that you publish your data with the Environmental Data Initiative (EDI) repository. Other data repositories are also acceptable - be sure to choose a repository that provides a DOI for the dataset you are publishing.

Data to be published: Publish only the subset of data used to perform the analyses published in your paper, at the coarsest level possible (e.g., include only the variables used in your published analyses and plot- or site-level mean values rather than all raw data). This is important to protect the many other ongoing projects within the network. Do not re-publish data that has previously been published.

Published dataset title: When publishing data in association with a manuscript, title the published dataset “Data from: (title of the associated manuscript)”.

Data authorship: As publishing data becomes increasingly common in ecology, and as the Nutrient Network grows, we have recognized that the list of manuscript authors and the list of people who contributed the data for these manuscripts are not identical. Thus, the goal of our data authorship policy is to decouple data authorship from manuscript authorship, and give credit to everyone who contributed data used in a NutNet publication, independent from contributions to manuscripts.

Attribution: Network-level leaders as well as everyone who was a site-level PI at a site included in your study during the years that the data were collected should be listed as an author of the dataset. Your paper would not have been possible without many hours of hard work by these people! Use the "NutNet-data-authors.csv" spreadsheet in the NutNet shared Dropbox or Drive to find this information. Authors should be listed in the “Creators” field in EDI.

Author order: Either Borer or Seabloom should be listed first and the other should be listed last. Remaining network leaders (Slette, Wilfahrt, Asmus, Lind, O'Halloran, Porath-Krause) and site leaders should be listed in alphabetical order. If you are publishing data from an add-on project, the add-on project leader(s) should be listed first, followed by remaining network leaders and site leaders.

Metadata: Include the following metadata for each site used in your study: site ID, site name, latitude, and longitude. Note that this is different from the “sites” field in EDI, which contains mostly LTER sites in a dropdown menu.

In the description, include the name of the data file(s) from the shared NutNet Data folder that you used for your analysis (e.g., full-cover_2024-05-11.csv).

Use the standardized NutNet metadata format, which is available in this file in the NutNet Data folder: nutnet-metadata-guide.csv.

In EDI, specify NutNet in the “Project” field.

Methods: Cite Borer et al. 2014 in the methods description.

Link to the associated manuscript: This should be done in the “Journal Citations” field in EDI.

Licensing of data: data should be published using CC BY, Creative Commons Attribution 4.0. This should be specified in the “Intellectual rights” field in EDI.

Step 10. When your manuscript has been published

Notify the NutNet Coordinator, who will list your paper on the publications page, and the Authorship Committee. We also encourage you to share the news with the full network by emailing the NutNet listserv <nutnet [at] googlegroups.com>.

Update the status of your abstract in the manuscript database on NutNet.org.

The process of leading a NutNet paper is outlined in the following flow chart:

Authorship Committee

Current members: Andrew MacDougall (University of Guelph), Lauren Sullivan (Michigan State University), Pedro Tognetti (University of Buenos Aires), Yann Hautier (Utrecht University), and Ingrid Slette (University of Minnesota)

Also: Elizabeth Borer and Eric Seabloom (University of Minnesota)

Former member: Carly Stevens

Current Network Coordinator

Ingrid Slette (slett152@umn.edu)