search

ternaustralia / ontology_tern

TERN Ontology
https://linkeddata.tern.org.au/viewers/tern-ontology
Creative Commons Attribution 4.0 International
6 stars 4 forks source link

What is the purpose of Sample geometry? #69

Closed nicholascar closed 2 years ago

nicholascar commented 2 years ago

Since a Sampling has a geometry - where it was done - we can infer that that's where a Sample comes from. What then is the value in recording a location for the Sample? Additionally, I think it misleading to record Sample location as the sample may actually change its place (i.e. when I take it back to the lab).

I think best to require location for the Sampling (as discussed: either directly with geometry or indirectly via links to a Site or indirectly via other spatial relations to a Feature) and to infer initial location for a Sample from that and to either discourage, or best prevent, location for a Sample. directly.

dr-shorthair commented 2 years ago

An act of Sampling does not have a geometry itself. It has a Feature of Interest, just like an Observation (and an Actuation) does. The Feature of Interest often has a location or geometry. The location or geometry of this Feature of Interest is likely to be the scientifically-interesting location or geometry for the Sample. So yes, I agree that it is the location of where it came from that is important, though the way this is modeled is not quite how you describe.

Mind you, a Site does have a location and geometry. And some Sites can also be thought of as Samples: for example, an Ecological Observation Site is usually positioned to be 'representative' of a tract or an ecosystem, for example. So the subset of Samples called 'Sites' (the Monitoring- and Observation-sites, anyway) probably do have an intrinsic geometry.

dr-shorthair commented 2 years ago

So I guess the proposal is to remove geo:hasGeometry from the expected properties of tern:Sample.

edmondchuc commented 2 years ago

Some survey protocols establish sites (tern:Site) where the site is further divided into quadrats or transects (tern:Sample). These quadrats or transects sometimes have a geometry and they relate to the tern:Site via sosa:isSampleOf. Please see this page for more information (yet to be migrated to the new site).

Different types of features (samples) are represented through the use of tern:featureType. It can be a quadrat, transect, plant individual, plant occurrence, animal occurrence, etc.

tern:Sample only defines the requirement of at least 1 value for both sosa:isResultOf and sosa:isSampleOf. It has access to geo:hasGeometry through its superclass tern:FeatureOfInterest.

tern:Sample: https://w3id.org/tern/ontologies/tern/Sample

nicholascar commented 2 years ago

OK, well then if the expectation is that Sample will usually only have a geometry by inheritance, i.e. by being an FoI, then I think that works well.

This again highlights a need to split Shapes from ontology, as per Issue #56, so that while ontological any old Site can have a geometry, the shapes, at least for NDES or BDR, will only be looking for them for special Sample types, such as Site, or for pure Sample and other kinds like MaterialSample looking for not them (Exactly 0).

edmondchuc commented 2 years ago

Ontology and SHACL are now separated as per https://github.com/ternaustralia/ontology_tern/pull/91.