The use of data from remote sensing - satellites, drones and similar tools - has grown steadily since the 1970s, supporting environmental assessment, resource management and scientific research. But simply having access to this data isn't enough: experimental satellite, drone and regional Earth Observation (EO) products often remain difficult to find, understand, reproduce and reuse. Their metadata, provenance, processing steps, quality information and access mechanisms are frequently incomplete, fragmented, or documented by hand. This gap is especially common for research-grade products that sit between raw satellite archives and mature, fully operational data services.
The OSCARS-funded project GOYAS - Geospatial Open-science Yielding ApplicationS, led by Fernando Aguilar Gómez at the Institute of Physics of Cantabria (IFCA, CSIC-UC), together with partners at the Institute of Marine Sciences (ICM-CSIC), the Institute of Marine Sciences of Andalusia (ICMAN-CSIC), and the Doñana Biological Station (EBD-CSIC), set out to close that gap - not by adding another documentation step at the end of the publishing process, but by making FAIR principles part of the technical design from the very start.
How it works
Data producers supply their data files along with a structured configuration file. From there, a pipeline built with Python and Snakemake (a workflow-automation tool commonly used in scientific computing) takes over, using GeoNetwork and GeoServer - established open-source platforms for managing and publishing geospatial data - to:
- Generate structured metadata using the ISO 19115/19139 standards, describing spatial and temporal coverage, variables, input data, processing steps, algorithms, parameters, units, uncertainty, quality information, provenance, licences and access links.
- Validate the metadata and assess it against FAIR criteria.
- Publish the data through standard geospatial access services (OGC-compliant web services, widely used so that GIS software anywhere can read the data without special adaptation).
- Assign a Handle persistent identifier (PID) - a permanent web link that keeps working even if the data is later moved or reorganised - and publish the final metadata record and landing page.
By folding metadata generation, validation, provenance capture and service publication directly into the ingestion process, GOYAS removes the need for data producers to manually manage complex XML metadata records or master every detail of geospatial metadata standards themselves.
What’s already published
The pipeline has already been used to publish several real Earth Observation products, including a chlorophyll-a concentration dataset, an inundation map for Doñana, a water-quality zoning product for the Mar Menor lagoon, and a sea surface salinity dataset - each now discoverable, citable and reusable through the platform.
Why it matters
For data users, structured metadata, provenance information and variable-level quality data (including precision and error values) make it easier to discover EO products, compare datasets, repeat analyses, validate results, and combine data from different sources - relevant to communities working with satellite-derived environmental data, climate studies, aquatic systems, land monitoring and geospatial analysis more broadly.
For data producers, GOYAS removes much of the manual burden of FAIR publication: a simpler, configuration-driven workflow handles metadata, validation, provenance documentation and PID assignment, letting researchers publish high-quality products without having to become geospatial metadata specialists themselves.
Rather than duplicating existing operational EO infrastructures, GOYAS is designed to complement them - supporting the experimental, regional and research-driven products that don't yet fit into large operational archives, while making them more visible and reusable. The team is now pursuing GOYAS's registration as an ENVRI-related EOSC service to support its long-term uptake and sustainability.