Environmental Information Australia - Supermodel

A "supermodel" is a set of models that have been designed to allow for the integration of data from different datasets. They achieve this by ensuring the models for each dataset implement the same patterns for common elements, such as spatiality, and they contain reference data, such as controlled vocabularies of values, that the datasets refer to, providing join points.

This Supermodel follows the generic Supermodel Model defined at https://linked.data.gov.au/def/supermodel.

Like all large and long-term data holders, the Department of Climate Change, Energy, the Environment and Water (DCCEEW) has many datasets that while they are conceptually related - within the environment domain - have been created separately and without interoperability as a priority. As a result, analysts must put lots of effort into aligning data before using multiple datasets together; effort that is likely duplicated by analysts unaware of others' work.

In early 2025, DCCEEW conducted a demonstration project called the EIA Supermodel Demonstrator that aimed at showing how integration-ready data might appear and be used.

The target datasets for this work were within DCCEEW’s Environmental Information Australia initiative and DCCEEW’s then new Biodiversity Data Repository was used as the reference dataset: the one whose form other datasets would emulate. This was due to the BDR having been designed specifically for cross-dataset integration.

The scope of this Supermodel was set within its establishment project to initially cover data from 6 DCCEEW datasets:

Additionally, the National Species List was included as a seventh dataset as it is already referenced by the BDR and species information is obviously used in much DCCEEW data:

An eighth dataset, Reference Areas was established too. This is a general-purpose, background dataset of reference spatial areas, such as RAMSAR wetlands and other named areas of ecological significance that are often used by DCCEEW for reporting:

This Supermodel contains Foreground Models for each of these datasets, as well as 20+ supporting assets - vocabularies, data validators - formulated according to the Supermodel Model specification. These are detailed below.

Together, the Foreground Models, Background Models and other Supermodel elements allow all parts of these DCCEEW datasets to be integrated, for example, the spatial parts of any two datasets may be overlayed, the observations from datasets that contain them can be used together and reference values for common attributes are aligned.

All the models in this Supermodel are implemented using the Web Ontology Language OWL. OWL is a very widely used, standardised, formal modelling language. Unlike UML models, OWL models natively have machine-readable forms allowing data made according to them to be automatically validated and processed into databases.

The BDR is natively modelled in OWL using the ABIS and AusTraits is too, using the OBOE, ETS and Darwin Core models. There is a pre-existing ANSIS OWL ontology too, into which it’s easy to transform data delivering service responses. The other 3 datasets - NVIS, SPRAT and HCAS - have had models, or partial model in OWL created for them, for the first time, within the project that generated this Supermodel

The Supermodel’s constituent models are related to one another using The Profiles Vocabulary PROF which provides properties to indicate when and how one model reuses - "profiles" - another. PROF also links models to validators created to test data claiming conformance to them.

The technical presentation of data modelled in OWL is in RDF which is a graph ('node-edge-node') data structure. RDF allows models and data according to the models to be stored in files or databases of the same sort. RDF is also infinitely extensible, allowing for not just model change and growth but for easy schema change and growth too.

Integrated use of multiple Foreground Models depends on each of them implementing, or mapping to, parts of the Background Models referred to as patterns. The patterns of relevance to this EIA Supermodel do not make a finite list as patterns can exist within and overlap other patterns, however the Patterns section lists a set that covers the scoped datasets' implementation at least once over.

For demonstrable interoperability, this Supermodel contains data validators for some of its Background and Foreground Models that test data claiming conformance to it against identified patterns. These validators are executable model specifications.

Most of the validators for the models within this Supermodel were created by the original model implementors, e.g. GeoSPARQL & ABIS, but one validator in particular was created for the EIA Demonstrator project: the EIA Data Profile. This validator is used to ensure datasets have basic metadata required to implement patterns listed in the Patterns section.

The Validators section below lists all the validators relevant to this Supermodel and indicates their dependencies. It also describes how to validate data.

Here is a list of terms and acronyms used in this document.

Like all the models within this Supermodel, the Supermodel itself has a machine-readable form that lists and relates all its elements:

Of the 7 Datasets in this Supermodel’s scope:

Only one dataset, SPRAT, is not yet bound in to the Supermodel by one of the methods applied to the other 6. This is planned to happen soon.

The AusTraits to NSL Mapping Linkset doesn’t need a Foreground Model either IF a single predicate can be added to either AusTraits.

Here are details of each of the four Foreground Model:

The listing of all Foreground Models used in this Supermodel is given in the EIA Test Models Catalogue where the models are indicated as having the role Foreground Model.

Models used to implement patterns as listed above but which are not Foreground Models aiming to model the entire contents of a Supermodel Datasets are Background Models.

Some of these Background Models are several steps removed from the Foreground Models where they are background models for other background models for the Foreground Models.

Many of these Background Models are well-known, international standard, models.

The listing of all Foreground Models used in this Supermodel is given in the EIA Test Models Catalogue where the models are indicated as having the role Background Model.