search

marco-bolo / eov-catalogue

The catalogue for MBO datasets intended as EOV contribution
0 stars 0 forks source link

Set semantic basis for BioEco EOVs #3

Open pbuttigieg opened 1 year ago

pbuttigieg commented 1 year ago

Working from discussions/outcomes from the GOOS BioEco Panel this week, and numerous ancillary discussions with GOOS OCG and other bodies, we will assist the BioEco Panel in drafting a semantic framework to stabilise the definitional space of the EOVs and their categories.

xref https://github.com/EnvironmentOntology/envo/issues/904

pieterprovoost commented 1 year ago

Can we use the BODC Parameter Usage Vocabulary semantic model here?

For example:

http://vocab.nerc.ac.uk/collection/P01/current/WWBMM00Z/ -> Wet weight biomass of phytoplankton per unit volume of the water body by optical microscopy and computation by morphometric method

Less granular:

http://vocab.nerc.ac.uk/collection/P01/current/SDBIOL04/ -> Wet weight biomass of biological entity specified elsewhere per unit volume of the water body

pbuttigieg commented 1 year ago

@pieterprovoost

We can map to it, but we first need to know what the semantics actually are. As we saw in the meeting, this isn't just a label matching exercise.

Further, this model is not a formal one - it's forcing expressivity that SKOS can't handle into an overloaded term.

pieterprovoost commented 11 months ago

Just a note that CSIRO have formalized the model in https://w3id.org/env/puv, but I'm also looking into OBOE.

marc-portier commented 11 months ago

just noting that BODC/NSV has some mappings from P01 to EOV (in A05) using the puv structuring indeed - but they seem to have abandoned that strategyin favour of the IADOPT approach

see https://github.com/vliz-be-opsci/nvsSPARQL/blob/feature/showcase-template-queries/templated-queries/02-eov-matching.ipynb for some testing on what they have available

for obvious completeness: their existing workings in no way limit our freedom to design, create and manage any external mapping - just adding this to make sure we don´t overlook what is already there.