The Supermodel

Overview

This is a documentation website for a Supermodel: a multi-part enterprise data model.

This particular Supermodel is for a public expression of the data holdings of the Geological Survey of Western Australia.

Question	Answer
What is a Supermodel, in detail?	See the next few sections See the generic description of a Supermodel
What is the status of this Supermodel?	See Status below
What other Supermodels?	See the Related Supermodels section of this Supermodel's description

An informal overview of this Supermodel and its various parts.

Introduction

This "Supermodel" is a multipart model used to coordinate expressions of data from multiple systems within the Geological Survey of WA (GSWA). Specifically, this model allows for:

• integration of data across systems
• multi-system data holding analytics
• system rationalisation planning
• delivery of coordinated/interoperable data

The major parts of this Supermodel are shown in Figure S1 above.

Only a few of the total Vocabularies and Component Models of this Supermodel have so far been defined: see the status in Overview/Status.

Related Supermodels

This Supermodel relates to several other Supermodels. The table below gives some details.

Supermodel	Relationship
Foundational Spatial Data Framework Supermodel	The FSDF Supermodel was created by Geoscience Australia to integrate multiple foundational spatial datasets within Australia. It specifies both a structure for spatial data and particular fundamental objects to join datasets with, such as authoritative boundaries for Australia and the states. This supermodel adopts the FSDF Supermodel ways of working for spatial data
Geological Survey of Queensland's Supermodel	A Supermodel for a similar agency to GSWA. It implements many of the same Background Models, Vocabularies and some of the same Component Models, such as Boreholes
Environmental Information Australia Supermodel	An emerging Supermodel that includes a series of national environmental datasets across several Federal government agencies. It shares many background models that cover sampling, sites, spatiality, projects etc.

Supermodel Structure

The structure of this Supermodel follows the Supermodel Model and consists of:

• Backbone Model
  • the high-level model that joins all the Component Models together
• Component Models
  • the detailed models for major datasets held by GSWA
• Background Models
  • models that the Backbone and Component models extend to implement generic patterns and keep aligned
• Vocabularies
  • controlled lists of terms used by all of the above models
  • the various models all indicate which vocabularies they rely on
  • see GSWA's vocabularies online

Additionally, this Supermodel contains the following model support content:

• PID Policy
  • a policy for the creation of Persistent Identifiers (PIDs) for all the models within this Supermodel and any data created to conform to these models

Technical Assets

All parts of this Supermodel, for example individual Component Models or supporting vocabularies, are outlined here in human-readable form (documentation) and their machine-readable forms for such as their schema and data validators are linked to as well.

Within the formal definition of this Supermodel , all the 'resources' of the Supermodel parts are listed and the roles that each play are given. See the Supermodel Definition.

Supermodel Definition

The machine-readable definition of this Supermodel is a Turtle file, online at:

• https://linked.data.gov.au/def/gswa-supermodel.ttl

Its content is rendered in human-readable form as follows:

Resource	Role	Description
Supermodel Definition	Profile	The formal definition of this Supermodel, including all its parts
Component Models
Bore Model	Component Model	A model that describes physical, functional and operational aspects of Bores, sometimes known as Wells.
Background Models
Vocabularies

Modelling Documentation

The following subsections apply to all models linked to and within this Supermodel.

Diagram Conventions

All the models within this Supermodel are visualised with informal OWL diagrams. These are diagrams that represent the Classes, Properties and Axioms of the Web Ontology Language with the elements shown in Figure K below.

Figure K: OWL Diagram Element Key

Namespaces

Namespaces provide unique identity to elements within this Model - classes, predicates, validation shapes and example data. Prefixes for namespaces are used to assist with documentation readability.

Where you see a prefix used, something like xxx:, it is to be replaced with the namespace for complete term definition. For example, using the table below, we can understand that bore:Bore is equivalent to https://linked.data.gov.au/def/borehole/Bore.

The following prefixed namespaces are used in class and property definition tables and the code examples following:

Prefix	Namespace	Description
ex	http://example.com/	Non-resolvable namespace for examples
bore	https://linked.data.gov.au/def/bore/	The namespace for the Bore Model
dcat	http://www.w3.org/ns/dcat#	Data Catalogue vocabulary: cataloguing international standard
dcterms	http://purl.org/dc/terms/	Dublin Core Terms: basic library catalogue-style metadata
geo	http://www.opengis.net/ont/geosparql#	GeoSPARQL: Semantic Web spatial data international standard
prov	http://www.w3.org/ns/prov#	Provenance Ontology: provenance data structures international standard
rdfs	http://www.w3.org/2000/01/rdf-schema#	RDF Schema vocabulary: Basic structural RDF elements
schema	https://schema.org/	The general-purpose schema.org model
skos	http://www.w3.org/2004/02/skos/core#	Simple Knowledge Organization System: a model for controlled vocabularies
xsd	http://www.w3.org/2001/XMLSchema#	XML Schema Definitions Datatypes

These namespaces appear at the start of RDF data files in the Turtle format (see next subsection) and SPARQL query text (the section after the next section) in a form similar to this table, for example in the schema for the Bore Model you can see the prefix borefor its namespace on the first line:

• PREFIX bore: <https://linked.data.gov.au/def/bore/>

RDF code

Many examples for the models within this Supermodel and all machine-readable assets are stored in the Turtle format. This format is a compact representation of RDF data and is generally considered to be the most human-readable form that raw RDF data can take.

An example:

ex:b-01
    a bore:Bore ;
    schema:depth [
        a schema:QuantitativeValue
        schema:value 239 ;
        schema:unitCode unit:M ;
    ] ;
.

In the example above (from the Bore Model's Quantitative Value class), we have a dummy ID for an instance of a Bore - ex:b-01 - with its class indicated - ex:b-01 a bore:Bore - and its depth given as a nested object, indicated by the predicate schema:depth:

[
    a schema:QuantitativeValue
    schema:value 239 ;
    schema:unitCode unit:M ;
]

The nested depth object above is a Blank Node, that is an object whose ID we don't need to know because it's a child object of something else.

All the prefixes in this code such as ex: and schema: are defined in the Namespaces section above and are expanded when this code is read by machine to full values, e.g.

ex:b-01 → <http://example.com/ex:b-01>

SPARQL queries

The SPARQL query language is the standard way to query Knowledge Graph's RDF data. SPARQL queries, which look a lot like SQL queries, specify a pattern that is used to extract part from a larger Knowledge Graph, a "subgraph".

For example, if we have the following data:

ex:b-01
    a bore:Bore ;
    schema:location ex:Kimberleys ;
    schema:depth [
        a schema:QuantitativeValue
        schema:value 239 ;
        schema:unitCode unit:M ;
    ] ;
.

ex:b-02
    a bore:Bore ;
    schema:location ex:Plibara ;
    schema:depth [
        a schema:QuantitativeValue
        schema:value 1073 ;
        schema:unitCode unit:M ;
    ] ;
.

ex:b-03
    a bore:Bore ;
    schema:location ex:Plibara ;
    schema:depth [
        a schema:QuantitativeValue
        schema:value 38 ;
        schema:unitCode unit:M ;
    ] ;
.

Then the following query may be applied:

SELECT ?bh
WHERE {
    ?bh 
        a bore:Bore ;
        schema:location ex:Plibara ;
        schema:depth/schema:value ?depthValue ;
    .

    FILTER (?depthValue > 100)
}

The above query will extract objects of class Bore - ?bh a bore:Bore - with location of ex:Plibara and depth greater than 100. The result will be the ID of ex:b-02.

Status

August 2024

This Supermodel is forming the backbone to the integration of multiple datasets in GSWA, such as MINEDEX & WAGIMS, that have not always been easy to crosswalk, as well as using up-and-coming national geosciene models, such as the Bores Model and the proposed GGIC Geochemistry Data Model.

Currently, as of August 2024, the individual models of this Supermodel that have been established are listed in the Component Models to the left.

License & Rights

This repository's content is available for reuse according to the Creative Commons Attribution 4.0 International (CC BY 4.0).

This content is copyright as follows:

© Government of Western Australia (Department of Mines, Industry Regulation and Safety), 2024

Contacts

This Supermodel is being authored for the Geological Survey of Western Australia within the Department of Mines, Industry Regulation and Safety.

For technical matters, please contact:

Nicholas Car
Geoscience Data Architect
Geological Survey of WA
Department of Mines, Industry Regulation and Safety
nick.car@dmirs.wa.gov.au
https://orcid.org/0000-0002-8742-7730

For other matters, please contact:

Geological Survey of Western Australia
https://dmp.wa.gov.au/Geological-Survey/Geological-Survey-262.aspx