The data was obtained from ALA. Here is the DOI to the data.

This repository contains a worked example of converting ALA data from Darwin Core to the TERN Ontology.

A mapping spreadsheet of the data to the RDF terms in the TERN Ontology.

The source data downloaded from ALA is the records-2021-12-01.csv file.

A mapping spreadsheet was created to map the columns of the CSV file to the TERN Ontology.

The run.py script is used to convert the CSV file to RDF.

• Mapped Faealla spreadsheet

View the data in Ontodia at https://ternaustralia.github.io/bdr-faealla-worked-example.

The worked example was created based on the following assumptions:

• Each row in the CSV file represents a Darwin Core record. These records are represented in the TERN Ontology with the class tern:RDFDataset.
• All items in a row (observation, sampling, etc.) are part of an tern:RDFDataset instance.
• The provenance and country code information are recorded in the tern:RDFDataset class as tern:Attribute instances.
• Sites are inferred from the location remarks and the decimal latitude and longitude values.
• Occurrences are recorded with the class tern:Sampling. The sampling events take place from within the established site. The result of the sampling is an occurrence recorded as a tern:Sample.
• The person who performed the samplings and observations also established the site.
• Three possible observations are made on the occurrence (fauna) sample:
  • sex (gender) of the occurrence
  • life stage of the occurrence
  • habitat description of the occurrence
• A specimen of an occurrence is recorded with the class tern:MaterialSample. The way the specimen was collected are recorded with the class tern:Sampling.
• Two possible observations are made on the specimen:
  • type status
  • taxonomic information (captured with the tern:Taxon)
• Attribution to people are recorded with the property prov:wasAssociatedWith.

TERN Support [email protected]