The role of domain repositories in sustaining high-quality data publications: researcher-oriented tools and strategies under limited resources

In the geosciences, most research results are supported by data. These data are measured, collected, generated or compiled by humans or machines (including numerical modelling) and they represent an increasingly important part of the research outcome. They should be made available and shared in openly in a reusable format wherever possible, while fully acknowledging the contributions of the individual researchers and institutions that collected or generated the data.

Research data repositories are permanent archives that provide access to data, metadata to related physical samples, as well as scientific software. An increasing number of repositories are assigning digital object identifier (DOI) to the data stored in their archives. The range of services offered includes fully self-service DOIs at large generic repositories, to institutional repositories that are open to institutional members only, and curated data publications by domain repositories specialising in data from a specific scientific field.

The involvement of skilled data curators, who are often also domain researchers, makes domain repositories the preferred destination for the publication of well-documented and reusable data. The generic metadata required for DOI registration is complemented by extensive, domain-specific metadata properties, such as the information on the temporal and geospatial domains, mineral or rock names, instruments and analytical methods. Ideally, this information derives from embedded controlled vocabularies or ontologies, which increase the discoverability of the data for humans and machines. During curation, author information is also supplemented with ORCID and ROR identifiers, and the published data is digitally connected to related research articles, datasets, software, and the physical samples from which the data were obtained. However, they are facing challenges due to insufficient staff to uphold these high publication standards. Unfortunately, the resulting delay in processing requests directs many researchers to generic repositories offering self-service DOIs that do not provide any data curation.

To address these challenges, GFZ Data Services provides intuitive tools for collecting rich metadata (metadata editors), data description templates with extensive explanations and online instructions on recommended file formats, for example. These tools enable researchers to provide high-quality metadata from the outset, thereby reducing the workload and time required for data curation.

In November 2025, GFZ Data Services launched ELMO, the fully revised and modernised version of our metadata editor. ELMO is not only a new web interface, but also contains many new features that improve the quality of metadata and the FAIRness of the data it describes, while simplifying the entry of information for researchers. For example, authors' names and institutions can be automatically entered by entering the ORCID; affiliations can be selected from a drop-down menu linked to the Registry of Research Institutions (ROR); and the controlled, linked data vocabularies already in use (e.g., GCMD and geosciML) are directly connected to the vocabulary services API, thus ensuring they are always up to date.

This presentation will outline the advantages and disadvantages of domain repositories, and introduce our new metadata editor ELMO.