Add semantics for CRS

[lvdbrink]

Official change request: OGC CR 598

Descriptions of Coordinate Reference Systems (CRS) are important in correct interpretation of spatial data, and in working with that kind of data. However, CRS descriptions are not readily available as Linked Data. Therefore, it would be good to have GeoSPARQL use a standard ontology for describing CRS, either as an extension or module of GeoSPARQL itself, or an endorsed external ontology.

[rob-metalinkage]

NB QB4ST is another place where CRS are referenced.

[mathib]

Hi!

Here are some additional notes regarding Coordinate Systems (CS') from use cases from the construction industry.

During the geometry modeling phase for construction works, a custom CS is often used and referenced to an agreed upon CRS. Either approach should differentiate between the applied Coordinate System (e.g. Cartesian CS/ Ellipsoidal CS, Custom CS/CRS, etc.), and propose relations between those coordinate systems, i.e. by defining transformations (translation, rotation, scaling). It might also be interesting to be able to define the units that are used in a coordinate system to allow meaningful analyses of distances, areas, and volumes, as required for quantity surveys.

This issue/extension proposal also has some overlap with Change 1: Decouple CRS and WKT (https://github.com/opengeospatial/geosemantics-dwg/issues/10).

A possible starting point for dealing with CS', transformations and (length) units in a wide set of use cases and modelling approaches is documented in the Geometry Metadata Ontology (GOM).

[jabhay]

MoSCoW poll created

[paulc-dstl]

EPSG maintain a registry of CRS and CRS components, transformations, etc. See for example the definition of British National Grid including the WKT specification. The EPSG codes are widely used so could form the basis of Linked Data PURIs.

[situx]

Today I have extracted all available proj4 definitions for another purpose. I figured why not converting them to RDF using the same process and try to figure out what would be necessary to create a GeoSPARQL extension ontology as proposed here.
The result is proj4rdf: https://github.com/situx/proj4rdf
It uses pyproj to extract features of the CRS and saves them in three literal types: WKT, Proj4, ProJJSON. It is a start and maybe we can expand upon it. CRS systems in this ontology model are represented as instances using the same EPSG URIs GeoSPARQL already uses in WKT literals

[dr-shorthair]

Is this encoding related to the ISO 19111 model? The CRS types appear to be the same as there.

[situx]

I am not really familiar with the ISO 19111 model.
What I know is:
EPSG Codes are unique identifiers for CRS and that proj4, projjson and WKT are formats that may express the parameters of a CRS.

[dr-shorthair]

ISO 19111:2019 is available from OGC here: https://docs.opengeospatial.org/as/18-005r4/18-005r4.html

[situx]

After reading the standard, yes these are the terms that are used in the software library as well and which - in any case - should be in an ontology model following this standard.

[lvdbrink]

This may also be of interest here: https://github.com/SEMICeu/epsg-to-rdf

From the README:

This XSLT is a proof of concept for the RDF representation of the OGC EPSG register of coordinate reference systems, extending the RDF mappings for reference systems defined in GeoDCAT-AP

[situx]

Thanks for the link, Linda. I did not know about that before. We better investigate what they have already done.