Linking services to enhance multidisciplinary dataset usability with semantic metadata in the Geo-INQUIRE Project

Geo-INQUIRE (Geosphere INfrastructures for QUestions into Integrated REsearch) addresses the pressing need for sustainable and interoperable research data infrastructures to tackle critical societal challenges. With a consortium of over 50 partners, the project integrates diverse data sources and services across geoscientific domains. By fostering FAIR (Findable, Accessible, Interoperable, and Reusable) principles, Geo-INQUIRE supports a cultural shift toward open and reusable science.

The complexity of modern societal challenges, such as climate change, resource management, and natural hazard mitigation, demands a multidisciplinary approach to research. Despite the importance of multidisciplinary research, significant challenges remain for users in accessing, integrating, and analyzing data across diverse scientific domains. The diversity of data formats, vocabularies, and structures often creates barriers to discovering and connecting related datasets, particularly when data stewardship services originate from different disciplines. 

In this contribution, we present a metadata approach designed to address challenges in data services interoperability and usability. A key challenge lies in improving the EPOS-DCAT-AP metadata model (https://epos-eu.github.io/EPOS-DCAT-AP/) to describe complex relationships between data services effectively. As of now, these are indeed unrelated, thus hindering users from connecting and querying related datasets. To address this, we introduced semantic information describing service input parameters and output response. 

The metadata approach is applied to a multidisciplinary use case involving two data services selected from the Geo-INQUIRE portfolio, provided by two different scientific domains: Anthropogenic Hazards and Geological Information and Modeling. The Anthropogenic Hazards domain provides access to data on human-induced seismic events in the form of a catalogue organized into episodes. Each episode consolidates industrial and geophysical data on anthropogenic activities in specific areas and in a given time period. 

The Geological Information and Modeling domain provides access to datasets on geological maps, boreholes (for water, oil, or gas extraction), 3D geological models, and mineral resources.

Specifically, linking seismicity data to borehole data allows users to build better subsurface models for more precise spatio-temporal analysis of relationships between industrial activities, subsurface geology, and seismic responses. For example, users can explore how specific borehole characteristics, such as lithology, correlate with seismicity patterns in a given region.

This work demonstrates how semantic metadata can significantly enhance the usability and interoperability of geoscientific data services. By explicitly defining relationships between services and leveraging metadata-driven automation, the approach significantly reduces users' efforts in multidisciplinary research.