Editor’s Note: Today’s post is by Nicola Davies. Nicola is Senior Publisher at IOP Publishing.

Open data is essential for transparency, reproducibility, and trust in research — key pillars of open science. While some funders, such as the National Institutes of Health (NIH) in the U.S. and UK Research and Innovation (UKRI) in the UK, have strengthened their policies to require open data sharing, many researchers are still not making their data publicly available. An analysis of over 30,000 Institute of Physics Publishing- (IOPP) published articles found that only one in ten researchers in the physical sciences share their data in a way that aligns with the FAIR principles—Findable, Accessible, Interoperable, and Reusable.

In environmental research, however, we are seeing a shift toward more open practices. Our recent analysis, supported by Dataseer, revealed that a growing proportion of environmental researchers are already sharing their data. Recognizing this momentum, IOP Publishing now requires researchers submitting to two of our environmental journals to share the underlying data for their papers. Not planning to share your data? No acceptance — unless there’s a compelling reason it can’t be shared. This move directly responds to the needs of researchers who are committed to sharing their data in a FAIR way alongside their published articles.

Why FAIR matters

The FAIR principles have been around since 2016 and are now widely recognized as the gold standard for data management. Developed by academics, librarians, and industry professionals, these guidelines ensure that research data is not just available but also well-organized and reusable. At IOPP, we actively promote these principles. We encourage authors to use standardized metadata, persistent identifiers, and clear usage licenses to ensure their research data can be easily located and properly attributed.

The evolution of our open data policy

Our mandate to disclose research data openly alongside submitted papers is the latest step in our journey toward greater transparency in scholarly publishing. In 2022, we required all authors to include a data availability statement in their articles. By 2023, we went further, asking authors who could not share their data to provide a justification. Now, our latest move — requiring open data for all submissions to Environmental Research: Food Systems (ERFS) and Environmental Research: Climate (ERCL) —is our most significant step yet toward greater transparency.

What are the risks?

There is always the risk that stricter requirements will deter submissions. Tsuyoshi Miyakawa, Editor-in-Chief of Molecular Brain, faced a similar challenge. In a 2020 editorial, he shared that when he requested raw data from 41 manuscripts, 21 were withdrawn, and 19 were rejected due to insufficient data. That means over 97% of those cases were unable or unwilling to present the raw data supporting their results — raising the question: did the data even exist? Of course, in some cases, there may be legitimate reasons for withholding data, such as concerns over patient privacy, intellectual property, or competition. But in environmental science, these constraints are typically less prevalent, making open data sharing both a feasible and necessary step toward greater transparency and reproducibility.

We’re eager to see how environmental researchers respond to our new policy. With data sharing becoming increasingly common in this field, we anticipate that the policy will align with this growing culture of openness. Will it encourage greater transparency and collaboration, or could it inadvertently discourage submissions? We’re hoping it will drive the field toward more rigorous, trustworthy science.

Recognizing legitimate barriers

Although we are committed to making scholarly communications more transparent, we recognize that not all research data can be shared openly and that other research communities are not yet in the same state of readiness for such a policy as in environmental science. Our recent analysis of publication data from more than 30,000 articles highlights that many researchers rely on third-party data — often owned by government agencies or private organizations — creating legitimate barriers to open sharing. We also learned that researchers’ perceptions and experiences vary across regions. In China and India, researchers are most likely to cite unfamiliarity with repositories as a key obstacle, whereas in the UK, concerns focus on data confidentiality and accessibility. In the US, the most common issue is that the data format may not be usable, even if shared, underscoring the need for better standards, tools, and a stronger culture of data management to enable wider access and interoperability. These barriers need to be addressed in a tailored and collaborative way. That is why our new data policy allows for exemptions where necessary and is only applied to two journals that serve a community that is already geared toward sharing their research data. We also believe that supportive policies tailored to the varied needs of research communities, greater awareness, and better training will be essential to making open data the norm rather than the exception.

As part of our efforts to facilitate data sharing, we have also introduced Dataset Articles in some of our environmental research journals. Dataset Articles provide structured, well-documented data that can be reused by other researchers. By providing detailed descriptions of data collection, processing, and potential applications, these articles enhance reproducibility, allowing other researchers to verify findings, build upon existing work, and generate new insights.

By strengthening our data-sharing requirements and offering options to share data in formats that suit the diverse needs of our research communities, we are advocating for better research practices. Change is not always easy, but we believe this step will help us to uphold the integrity of scientific publishing and create a potential launchpad for extending our policy to other journals too.