Speaker
Description
Ecological research infrastructures increasingly require environmental context to support ecosystem experiments, biodiversity analyses and cross-site syntheses. Open satellite archives and cloud platforms now provide unprecedented access to Earth Observation (EO) data. However, these products often remain difficult to integrate into ecological workflows because they require substantial technical expertise, implicit modelling assumptions and dedicated infrastructure services to become ecologically interpretable and reusable.
We propose an accessible framework for transforming EO products into analysis-ready, FAIR and semantically interoperable ecological variables. The framework comprises four components: an EO computing platform, cloud storage, Zenodo, and Data Management Centres (DMCs). Within this workflow, we introduce semantic FAIRification as a complementary layer to existing geospatial metadata standards (e.g. STAC, OGC/ISO), making explicit the provenance, ecological interpretation, uncertainty, and modelling context of EO-derived variables.
Sentinel-2 phenological metrics provide a practical example. Annual NDVI trajectories can be classified into phenological classes, from which metrics such as phenological diversity, evenness, and edge density can be derived. These variables may represent candidate proxies for habitat heterogeneity, resource continuity, or temporal ecosystem variability. Importantly, semantic annotation does not validate ecological mechanisms, but formalises transparent and testable ecological hypotheses.
An ongoing Danish case study based on long-term insect occurrence data illustrates how Sentinel-2 temporal metrics can be connected to ecological observations through reproducible workflows and semantic documentation. The study is presented as a demonstrative integration framework, rather than ecological validation, showing how EO-derived variables may become interpretable, reusable and testable components of biodiversity inference.
We argue that the next frontier for EO in ecological research infrastructures is not producing more satellite products, but making their meaning, assumptions and reuse conditions explicit. Semantic FAIRification may transform EO products from downloadable raster layers into reusable ecological variables and future DMC-enabled services.
Keywords: Earth Observation, FAIR Data, Semantic Interoperability
| Are you participating to the "AnaEE Environmental Rising Star Award "? | No |
|---|