Methodology for building interoperable Research Infrastructures: Example from Geo-INQUIRE project

Building and establishing a fully interoperable Research Infrastructure (RI) allowing smooth data exchange across various scientific domains is a dream of many researchers and data managers around the world. Indeed, there are various pathways to achieve it and many attempts are currently being made. Unfortunately, there is no single approach to fit all RIs since there is natural heterogeneity that encompasses the different phases of the data lifecycle, starting from data collection until data interpretation spreading across various domains. 

In the current contribution we present an example methodology for building an interoperable RI which is being applied within the Geosphere INfrastructures for QUestions into IntegratedREsearch project (Geo-INQUIRE, https://www.geo-inquire.eu/). Geo-INQUIRE was launched in October 2022 and comprises a unique consortium of 51 partners, including national research institutes, universities, national geological surveys, and European consortia. A portfolio of 150 Virtual Access (VA) and Transnational Access (TA, both virtual and on-site) installations will be offered to the scientific community  across domain barriers, especially the land-sea-atmosphere environments, including EPOS, EMSO, ARISE, ECCSEL, and ChEESE RIs. The great challenge is to find common grounds across the domains and define principles general enough for all participating RIs, though detailed enough allowing useful interoperability. The example methodology we are presenting here has been developed within implementation of European Plate Observing System (EPOS; an European Research Infrastructure Consortium since 2018) and relies on data access provision through web-services (APIs). For achieving true interoperability it is not enough to have robust and efficient web-services but also having rich metadata description provided by sufficient metadata models compliant with FAIR principals is critical. The methodology therefore envisages data integration through an approach that includes web-services, rich metadata and semantics. These three key elements are monitored and evaluated through a set of criteria distributed into three levels which were put together into an Implementation Level Matrix (ILM) to understand the landscape of service provision. This ILM serves as a tool for capturing the changing maturity of installations/services and tracking their readiness for interoperable integration into RIs.

The interoperability starts at the data provider level and therefore domain-specific coordination and common development efforts are important.  A good example of this is the work done in seismology. After a long process of joint work with experts in data FAIRness a group of data centres submitted to the International Federation (FDSN) a proposal to update their community guidelines about DOIs for seismic networks. This was iterated, discussed, and finally adopted as a community standard in December 2023. However, the guidelines include not only topics exclusive to seismology, but also the aim of being as FAIR as possible from the point of view of a multi-disciplinary perspective. The challenge to extrapolate this to the RIs and data providers from other disciplines taking part in the project and adapt them to their reality is an on-going effort within the project.