OGCGeoSPARQL is the Open Geospatial Consortium's standard for representing and querying geospatial data in RDF. By providing a small but powerful ontology combined with SPARQL extension functions, it enables spatial queries and reasoning over linked geospatial data, bridging the Semantic Web and geographic information systems communities.
Background
GeoSPARQL emerged from the OGC's efforts to bring geospatial capabilities to the Semantic Web. The standard was submitted to the OGC by a consortium that included the Australian Bureau of Meteorology, CSIRO, Ordnance Survey, Oracle, USGS, and several other organizations. Version 1.0 was published in 2012. After nearly a decade of increasing adoption, the GeoSPARQL Standards Working Group was re-established in September 2020 to develop a non-breaking update. GeoSPARQL 1.1 was published in 2024, adding SHACL validation, a Simple Features geometry class hierarchy, and an expanded function vocabulary.
Purpose & Scope
GeoSPARQL provides three interconnected components:
- A topological ontology in RDFS/OWL for representing Features and Geometries
- Geometry literal serializations supporting WKT, GML, GeoJSON, KML, and a placeholder for DGGS (Discrete Global Grid Systems)
- SPARQL extension functions for spatial querying, covering topological relationships (Simple Features, RCC8, DE-9IM) and geometric computations
These components work together to allow SPARQL queries like "find all features within a given bounding box" or "return all features that overlap a specified region."
Topological Vocabularies
| Vocabulary | Description |
|---|---|
| Simple Features | sfContains, sfWithin, sfOverlaps, sfTouches, sfCrosses, etc. |
| RCC8 | Region Connection Calculus relationships (8 relations) |
| DE-9IM | Dimensionally Extended 9-Intersection Model (Egenhofer/Clementini) |
Serializations & Technical Formats
The GeoSPARQL ontology itself is published in RDF/XML and Turtle. Geometry data within GeoSPARQL uses WKT literals, GML literals, or GeoJSON literals embedded as typed RDF literals.
Implementations
GeoSPARQL has been implemented (at least partially) in numerous triple stores and geospatial databases including Apache Jena (since v2.11), Eclipse RDF4J, GraphDB, Stardog, Virtuoso, MarkLogic, Oracle Spatial, QLever, and Ontop VKG. Compliance and performance benchmarking have been conducted through the Geographica benchmarks and a dedicated GeoSPARQL Compliance Benchmark developed by Jovanovik et al. (2021).
Governance & Maintenance
GeoSPARQL is maintained by the Open Geospatial Consortium through its GeoSPARQL Standards Working Group. The standard's development is open and can be followed on GitHub. The GeoSemantics Domain Working Group coordinates broader efforts at the intersection of geospatial data and semantic technologies.
Notable Implementations
The Ordnance Survey Linked Data Platform uses GeoSPARQL-compatible properties. The LinkedGeoData project (AKSW, University of Leipzig) represents OpenStreetMap data using GeoSPARQL vocabulary. Government geospatial data portals in the UK, Ireland, and Australia have adopted GeoSPARQL patterns.
Related Standards
- SPARQL -- The query language that GeoSPARQL extends
- OGC Simple Features -- The feature model underlying GeoSPARQL's spatial predicates
- GML -- One of the supported geometry serialization formats