A new article type: The 'Data Article'

Without data, there can be no science. Most research papers in the European Journal of Soil Science (EJSS) describe their datasets and how these were used to reach the paper's results and conclusions. These datasets are sometimes re-used or integrated with other datasets by the original researchers or selected collaborators. But this is far from the full value that could be extracted from a dataset, if the ingenuity of other researchers, perhaps combined with other datasets, could be applied to the dataset.

The datasets themselves are important scientific products, worthy of a detailed description of their sources and the methods used to reach a final product, independent of the original research questions for which they were collected. Well-documented datasets allow other researchers to understand their strengths and limitations, and to judge how they might be used to advance their own research. Further, meta-analyses and studies comparing various datasets depend on reliable, well-documented, and accessible datasets.

Nowadays, scientists are being motivated, often as required by funding bodies, to adopt Open Science principles, that is, to make scientific research accessible throughout society and thus enhance knowledge dissemination, leading to greater scientific and societal impact. Open Science includes the FAIR data principles: datasets must be Findable, Accessible, Interoperable, and Reusable. This movement is generally supported by scientists as well as the European Commission, which has developed a European Data Strategy incorporating FAIR concepts. Researchers need to be able to publicize their datasets (“Findable” and “Accessible”) and share them (“Interoperable” and “Reusable”) while maintaining a high standard of scientific publication through peer review in scientific journals.

Therefore, the EJSS has decided to add a new category of articles: the Data Article. Although there are journals solely focusing on data, EJSS data articles are differentiated by their direct relevance and explicit links to soil science research, that is, applications that could be the subject of other article types published in the EJSS and other soil science journals.

Data articles should give sufficient information to enable other scientists, whether soil scientists or those in related disciplines, to properly use the datasets in their own research. In line with the European Data Strategy, this will also enable readers outside of science, including industries, land managers and policymakers, to use datasets published in the EJSS. Especially for the non-specialists in soil science research, the datasets must be well-described so that they will be properly used.

An important part of this is that the datasets must be accessible in digital form and in standard formats. They must be made available under a defined licence and on-line access procedure, including a Digital Object Identifier (DOI). Access is preferably non-exclusive and without registration, but if required by applicable law (e.g., privacy regulations), under well-defined and non-discriminatory procedures for access (e.g., a Memorandum of Understanding). Thus, the Data Article category is not suitable for proprietary datasets used in closed research projects -- these must be described in an Original Manuscript or Short Communication.

The structure of a Data Article differs considerably from other article types. In the same way that a research paper's methods should be reproducible, the methods by which a dataset was produced, and the assumptions made in this process, should be transparent. Articles in this category should describe the workflow by which a dataset was produced, from the collection of the raw data or other sources through to the finished product(s). The descriptions in the paper should be sufficient to enable other scientists to create similar datasets with the same procedures. For example, what was the sampling strategy and field or lab protocol, how were the datasets cleaned or harmonized with other datasets, were outliers removed and on what grounds, how were they evaluated for completeness and correctness, and how was uncertainty or accuracy evaluated? In the case of compilations or manipulations of primary datasets, how was this done, and how were the diverse datasets harmonized? In the discussion, the presented dataset(s) may be compared and contrasted with similar products. Other topics that can be addressed in the discussion may include, for what applications are the data suited, for what applications have the data already been used, and what are the limitations of the data and the potential for improvement?

We hope that with the introduction of this new article type the EJSS will be able to accept contributions of datasets of high relevance and scientific importance to soil science broadly defined, independent of their use in specific research projects, in compliance with FAIR data principles.