Integrating biodiversity in Situ data, Earth observation and stakeholder engagement - from machine- to policy-actionability

The EU’s Biodiversity Strategy for 2030, a core part of the European Green Deal, addresses the complex relationship between human society and its environment by prioritizing the restoration of ecosystems and building resilience against climate change, deforestation, and biodiversity loss.

These environmental stressors do more than just degrade ecosystems; they create a pressing need for policymakers, researchers, and society to actively track and mitigate ecological shifts. In order to design effective mitigation strategies, new political frameworks and massive simulation infrastructures are being developed with the aim to establish a common European Green Deal Data Space. The involved initiatives rely on the integration and standardization of diverse, large-scale datasets, ranging from long-term biodiversity records (e.g., eDNA) to real-time IoT sensor data (e.g., camera traps) and global Earth observation (EO) data combined with model-derived reanalysis datasets like ERA5.

‘Biodiversity Meets Data’ (BMD) is a Horizon Europe project delivering a unified access point for AI-assisted biodiversity monitoring and cross-realm (terrestrial, marine, freshwater) analysis tools representing a key contribution to the thematic expansion of the European Green Deal Data Space ecosystem. By providing a robust technical infrastructure, BMD facilitates the quantification of diverse ecological pressures - ranging from climate change to land-use shifts - on biodiversity. The project is strategically focused on the EU Natura 2000 network, equipping stakeholders such as conservation managers and policy makers with the necessary tools to implement and evaluate EU Nature Directives such as the Birds and Habitats Directives.

In this talk, we will present how BMD leverages FAIR Digital Objects (FDOs) and data space concepts around governance, licensing, and provenance tracking to synthesize computational workflows and diverse datasets into actionable knowledge units (“Workflow Run RO-Crate”, Figure 1). We will demonstrate our implementation path for such data-rich, self-contained digital containers building on web-based technologies such as RO-Crate (lightweight data packages) and FAIR Signposting (machine-interpretable layer describing resources). Those webby FDOs are designed to bridge the gap between practical needs of conservation stakeholders such as supporting data-driven decision making and technical capabilities of the Green Deal Data Space ecosystem.

Integration of targeted feedback from stakeholders, notably Natura 2000 site managers, into our development process ensures that the FAIR-compliant data products and FDO service framework are not only technically robust, but also socially and politically actionable.

 

Figure 1. Throughout its life cycle in the BMD data space, data is represented as RO-Crate. Initially (left), the data and the computational workflow are bundled as Workflow RO-Crate. Following processing, this is combined with the results and enriched with retrospective provenance and metadata to form a Workflow Run RO-Crate (right). Finally, these are presented as webby FAIR Digital Objects, incorporating a machine-interpretable layer based on FAIR Signposting (bottom).