W3C home > Mailing lists > Public > public-sdw-wg@w3.org > May 2017

Implementations of SOSA at GA [SEC=UNCLASSIFIED]

From: Car Nicholas <Nicholas.Car@ga.gov.au>
Date: Wed, 17 May 2017 13:57:08 +0000
To: "public-sdw-wg@w3.org" <public-sdw-wg@w3.org>
cc: "Kemp Carina" <Carina.Kemp@ga.gov.au>, "Butcher Stirling" <Stirling.Butcher@ga.gov.au>, "Ip Alex" <Alex.Ip@ga.gov.au>, "Bastrakova Irina" <Irina.Bastrakova@ga.gov.au>
Message-ID: <1495029426606.10759@ga.gov.au>
Dear SDW WG,

I am writing to let you know of some implementations that Geoscience Australia (GA) has made of your SOSA ontology.

1. Samples Register - SOSA Samples

​GA delivers a Samples Register (Catalogue) (http://pid.geoscience.gov.au/sample/) through which the metadata for 3M+ physical samples is delivered. The  Register is presented as a Linked Data API that delivers metadata according to several ontologies, one of which is the SOSA ontology view.

For the sample AU239, the persistent URI for it is http://pid.geoscience.gov.au/sample/AU239 and that uses the IGSN ontology for its default view. The SOSA ontology view of it, in turtle format, is available at http://pid.geoscience.gov.au/samples/sample/AU239?_view=sosa&_format=text/turtle. Every sample has this view so you could create a SOSA-only view of our entire Samples Register by crawling the paginated register for all sample identifiers and appending the query string args _view=sosa&_format=text/turtle to each link.

In order to create this view I used the documentation of the SOSA ontology available at https://w3c.github.io/sdw/ssn/#Sampling, particularly the images such as Fig. 10. Reading the basic descriptions of the main classes such as Sample and FeatureOfInterest and viewing the various diagrams, I was able to fairly easily implement some simple RDF that was similar to my previous IGSN ontology implementation. I will be deprecating the IGSN ontology's  own classes and relationships in favour of those in SOSA thus soon SOSA, with some SKOS classification additions, will become our Samples Register's default ontology.

2. Surveys Register - SOSA Sampling

GA delivers metadata for "surveys" - acts of aerial and ship-borne magnetic and radiometric surveying via our Surveys Register, available online at http://pid.geoscience.gov.au/survey/, which is similarly a Linked Data API with multiple ontology views.

The SOSA view, in turtle format, of Survey 801 is at http://pid.geoscience.gov.au/survey/801?_view=sosa&_format=text/turtle. Like for the Samples, all Surveys have a SOSA view. Note also that the TIME ontology's constructs for a temporal event's start and end times are used here.

A GA survey is interpreted as a sosa:Sampling event. While no link to resultant Samples are given in the Surveys Register information, our dataset register will link back to these Surveys and soon the total graph of Surveys and their resultant Samples' data will be cached in an RDF triplestore and made available publicly. An example of a Survey's dataset is in our public dataset catalogue at http://ecat.ga.gov.au/geonetwork/srv/eng/search#!e9e9c7f7-6896-384f-e044-00144fdd4fa6 but this is not yet delivered in RDF or with SOSA semantics or linked to its generating Survey.

3. Future Work: Entities Register - SOSA Samples and related Samples

GA maintains a register of "Entities". These are sampling locations such as boreholes. We will shortly be publishing this as yet another Linked Data API like Samples and Survey and linking physical samples to them. When we do (ETA 1 month) we will publish a SOSA view as the default view and ensure that our existing Samples Register and this register are harmonised. Currently the Samples API indicated an Entity as a "site" via a SOSA sampling hasSampleRelationship ​but with no valuable natureOfRelationship value. We may have to implement a vocab of terms for natureOfRelationship such as subsample unleess SDW WG has a better suggestion.

This Register is a cornerstone register for GA with lots of information linked to it so once implemented in SOSA, many other registers of data will eventually adopt SOSA too.

4. Future Work: QB4ST in netCDF

GA delivers lots of spatial data in netCDF containers, such as the datasets resulting from Surveys. Currently that data is in "normal" netCDF files but we are already starting to implement both SKOS and PROV ontology terms within the netCDF headers. Through the work of the informal netCDF-LD WG which GA and CSIRO are members of, we hope to soon implement Linked Data metadata in netCDF files more generally. When we do, we will most likely implement QB4ST descriptions of variables. ETA for this, 3 - 6 months.


At GA we are interested in assisting with the progression of SDW WG outputs towards W3C Recommendations. To this end, we will assist with further ontology adoption where we can (have the data to do so). Please let me know if there are other datasets or metadata that we have that we could apply SDW WG semantics to.



Nicholas Car

Data Architect

Geoscience Australia

Geoscience Australia Disclaimer: This e-mail (and files transmitted with it) is intended only for the person or entity to which it is addressed. If you are not the intended recipient, then you have received this e-mail by mistake and any use, dissemination, forwarding, printing or copying of this e-mail and its file attachments is prohibited. The security of emails transmitted cannot be guaranteed; by forwarding or replying to this email, you acknowledge and accept these risks.

Received on Wednesday, 17 May 2017 13:57:46 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 17 May 2017 13:57:47 UTC