MOC Open Science Data and Knowledge Graphs
Research data is only reusable if someone who was not there can interpret it. That is the whole subject of this map, and the notes below are layers of one stack: a container format, conventions giving its contents meaning, community profiles making those conventions checkable, and identifiers letting the result be cited and traversed.
The layering matters because each layer is useless without the one beneath it. netCDF specifies how to attach metadata but not what to attach; CF conventions supply the semantics; ATMODAT and the UC2 data standard turn those into a mandatory, machine-checkable list. “It is netCDF” is a claim about the container and not about interoperability, and most real exchange problems live in that gap.
Above the file sits the question of what the file is: metadata describes it, provenance records what produced it, DOIs make it citable, and FAIR is the framework tying those together. Above that again, knowledge graphs and ORKG attempt to make the relationships between research objects queryable rather than merely readable — with MaRESS and machine-actionable knowledge for Earth science as the domain-specific efforts, and ORKG reborn and the NFDI4Earth keynote as notes from the surrounding community.
Core notes
- FAIR
- Metadata
- Provenance
- Knowledge graph
- ATMODAT
- UC2 data standard
- netCDF
- CF conventions
- common data language
- climate variable
- climate data
- Open Research Knowledge Graph
- ORKG reborn
- DOI
- Machine-actionable Knowledge for Earth Science
- NFDI4Earth plenary keynote
- MaRESS
- Research Rabbit
- AMIP
- spatial reference system