Essential Variables as a semantic interoperability solution for the Green Deal Data Space

The Green Deal Data Space is born in the big data paradigm where many sensors produce constant streams of Earth observation data (remote sensing and in-situ). The traditional and manual organization of data in layers is no longer efficient as data is constantly evolving and combined in new ways. There is a need for a new organization and representation of the data. Data Spaces are intended to become the EC’s comprehensive solution to integrate data from different sources with the aim to generate and provide a more ready to use knowledge in support of the Green Deal priority actions on climate change, circular economy, pollution, biodiversity, and deforestation.

Semantics need to be moved from the layer level to the property level. The solution is not a comprehensive new data model, but a framework composed by a suite of ontologies implemented in line with best practices, reusing existing standard vocabularies, such as Essential Variables (EVs). These are used in Earth observation to define variables that correspond to high impact on the Earth systems and should be a priority for monitoring. EVs assume that there is a limited number of variables that are essential to characterize the state in a system without losing significant information on its past and future trends.

In AD4GD (Horizon Europe nº 101061001), we are describing EVs following the I-ADOPT ontology, starting with the Essential Biodiversity Variables, and extending the model to describe the concept of EV products, where products define spatial and temporal resolution constraints. I-ADOPT (Barbara Magagna et al, RDA) is an ontology framework designed to facilitate interoperability between existing variable description models across research domains. It provides a common set of core components and relations to define machine-interpretable variable descriptions that re-use FAIR (Findable, Accessible, Interoperable, Reusable) vocabulary terms.

I-ADOPT defines 5 classes (Variable, VariableSet, Property, Entity, and Constraint), and specifies several object properties. Thus, i-ADOPT enables the decomposition of complex observable properties into essential atomic parts represented through concepts in FAIR terminologies, serving as a common layer of abstraction to systematically align and extend concepts from different terminologies as needed.

EVs expressed under this new ontology schema are being translated as linked data to become available via the OGC RAINBOW Definition Server (https://defs.opengis.net/vocprez/). RAINBOW will ensure that Essential Variables and Products are assigned unique and persistent identifiers facilitating wider adoption and reuse.