LUMEN Deliverable 4.2: LUMEN Data Model

Scientific discovery today remains fragmented across domains, platforms, and disciplines. D4.2 - LUMEN Data Model addresses this fragmentation by providing a shared semantic and governance framework enabling scientific communities to present their outputs consistently and to support cross-disciplinary discovery. Key terms such as Scientific Community, Scientific Resource, Data Product, Data Contract and Data Mesh are clarified in Section 0. Definitions to support a smooth reading experience. The model implemented in this task builds directly on the architectural vision of D4.1 (Homo et al., 2025) and translates it into an operational specification. It defines a lightweight but rigorous metadata model, a data governance model, and a semantic exposure model that allow communities to make their outputs machine-actionable and interoperable. At its core are two foundational entities: (i) Data Products, which represent what a community exposes to the LUMEN wider ecosystem; and (ii) Data Contracts, which define how these products are governed, accessed, and stewarded. Instead of modelling technical access points as independent entities, D4.2 treats exposure mechanisms, such as APIs, endpoints, and distributions, as integrated properties of Data Products, all governed through their associated Data Contracts. This streamlined approach reduces complexity, ensures alignment with standards like DPROD, DCAT, and ODCS, and keeps the model focused on community-facing concepts rather than underlying infrastructure. The model advances the interoperability objective through three complementary features. Shared and standardised metadata fields ensure that exposed scientific resources are findable, citable, and compatible across domains. Support for disciplinary standards enables communities to represent their research nuances, so that, for instance, a geospatial dataset, a molecular simulation, or a software record retains its precision. Governance structures connect products to people, institutions, and projects, enabling transparent attribution, version management, and community-driven evolution. The benefits of this standardised exposure model are immediate. Meta-Search services can align scientific outputs more consistently; contractual conditions become explicit and machine-readable; and integration costs decrease, as communities expose their outputs once and make them discoverable across the entire ecosystem. This deliverable is currently under review by the EC.