What happens when FAIR is built in from the start? Insights from the GOYAS Project

The Geospatial Open Science Yielding Applications (GOYAS) project, developed under the Horizon Europe OSCARS framework, demonstrates a comprehensive pathway from FAIR principles to operational practice for Earth observation (EO) data products. While the FAIR principles (Findable, Accessible, Interoperable, Reusable) are widely endorsed by research data communities, translating them into reproducible and scalable workflows across heterogeneous data providers remains challenging. This contribution presents concrete results and lessons learned from GOYAS project, which has developed and implemented a FAIR-by-design system that supports community adoption and cross-disciplinary data reuse.

At its core, GOYAS comprises a set of customized software components, including an automated data production pipeline, a georeferenced data repository, and an OGC-standard API endpoint. The data ingestion pipeline integrates automation that reduces the initial effort required from data producers to generate FAIR data, by automatically producing standardized metadata, provenance information, and quality metrics as a by-product of routine processing. This approach enables transparency, consistency, and long-term reuse across all stages of the data lifecycle. To enforce the “F” of Findability, persistent identifiers (PIDs) are minted for mature data products using EOSC-Beyond services, ensuring persistent, machine-actionable references and reliable data product traceability.

A key outcome of GOYAS is the implementation of a validation framework that acts as a prerequisite for the publication of final data products, whereby persistent identifiers are assigned only to validated outputs. Each product undergoes:

• Metadata standard validation, ensuring compliance with agreed schemas and machine-readability requirements (ISO 19139);

• INSPIRE alignment, verifying that spatial data components meet European geospatial interoperability standards;

• FAIRness evaluation using FAIR EVA (Evaluator, Validator and Advisor), assessing the degree to which products comply with FAIR principles through automated tests.

Only when all validation checks are successfully passed is a product considered mature for publication and assigned a persistent identifier (PID), thereby guaranteeing discoverability and long-term referenceability within EOSC and beyond.

We discuss how FAIR-by-design principles were embedded at key architectural layers, including metadata generation, PID minting, and automated quality assessment, and how these design choices support not only technical interoperability but also community adoption. Lessons learned highlight the importance of early integration of FAIR requirements into workflow design, the practical challenges of harmonizing cross-domain standards (FAIR and INSPIRE), and the role of automation in enabling scalable FAIR implementations without imposing additional effort on data producers.

By providing a documented and operational model that combines FAIR principles, persistent identification, standards compliance, and automated validation, GOYAS advances the practical implementation of FAIR and open data management in environmental sciences and offers transferable insights for related research communities.