Monitoring soil health at regional and global scales depends on the ability to integrate and harmonise soil data collected across institutions, laboratories, and monitoring programmes. Despite increasing volumes of soil information, datasets remain highly fragmented and are often neither publicly accessible nor easily discoverable. When available, they are typically stored in heterogeneous formats and frequently lack consistent metadata or methodological descriptions. These barriers limit data reuse and constrain the development of effective soil monitoring systems.
The Varda Foundation, an independent non-profit stewarding digital public goods for agriculture, addresses this challenge through SoilHive, an open digital platform designed to support the discovery, harmonisation, and reuse of soil data. Rather than generating new datasets, SoilHive focuses on making existing data work together. It is built as precompetitive tool: a shared, open infrastracture on top of which researchers, institutions, and developers can train new models, and build new data products. In September 2026, SoilHive will be released as fully open-source, with a modular architecture designed for federated deployment across institutions. This is not just a technical choice, it reflects a commitment to building something the community owns and shapes together.
At the core of the platform is a robust data harmonisation module designed to directly tackle the fragmentation problem described above. Users can bring heterogeneous datasets into a common structure through a guided interface, without requiring deep technical expertise. The underlying data model covers more than 50 core soil properties and 80 sub-properties, structured through explicit parent–child ontological relationships, with controlled vocabularies, and standardised units with automated conversion. Once harmonised, data is immediately model- and AI-ready, accessible via both interface and API, and structured for direct integration into downstream analytical and machine learning pipelines.
To ensure the platform remains relevant across different geographies and user communities, new functionalities including interpretation dashboards and data visualisation tools are being co-developed with a diverse range of stakeholders, from research institutions and policy makers to farmer organisations, across Europe and sub-Saharan Africa.