CAMELS-GB in RDF: Catchment Attributes and Meteorology Dataset for Great Britain as a Knowledge Graph

This dataset presents an RDF (Turtle syntax) version of the CAMELS-GB (Catchment Attributes and MEteorology for Large-sample Studies – Great Britain) dataset, generated from CSV files using the open-source HydroTurtle tool (https://github.com/shamilasudalshana/NFDI4Earth-HydroTurtle). The purpose of this conversion is to enhance the interoperability, reusability, and FAIRness of the hydrometeorological data, in line with the principles of the Semantic Web and Linked Open Data.

The dataset covers the period 1970–2015 and includes data for all 671 catchments from the original CAMELS-GB dataset. Attributes span multiple domains including climate, hydrology, topography, land cover, and soil properties. Some attribute categories are pending inclusion and will be added in future updates.

The RDF data modeling is based on a combination of established ontologies, including:

• SOSA/SSN (Sensor, Observation, Sample, and Actuator)

• QUDT (Quantities, Units, Dimensions, and Data Types)

• GeoSPARQL (for spatial data)

• Dublin Core, Schema.org, and SKOS (for metadata and classification)

• EnvThes (for environmental terminology)

• HM-Ontology – a custom ontology developed as part of this project, to cover hydrological and meteorological concepts not addressed by existing vocabularies.

URIs are assigned to individual catchments, variables, and observations, enabling integration with other Linked Data sources and facilitating semantic queries. The RDF output is organized per catchment as multiple Turtle (.ttl) files. When loaded into a triple store, these can be unified into a single graph for querying and reasoning.

This conversion was developed as part of the NFDI4Earth inititative. The dataset is licensed under CC-BY 4.0.

This dataset is intended to support researchers and developers working on hydrology, environmental data integration, knowledge graphs, and natural language question answering over Linked Data.