search

nvs-vocabs / OBISVocabs

A repository for the management of issues related to vocabularies used by the OBIS community.
2 stars 0 forks source link

eMoF structure question #28

Open roswri opened 2 weeks ago

roswri commented 2 weeks ago

The BODC vocab management team have received a query about how to structure data about marine species washed up on the beaches. Individuals are found in many different forms (alive, dead, only a shell or exuvia) and all this information is noted. Many of the measurement values they would like to use are in the S12 vocabulary, but no there is no appropriate measurmentType value in P01 currently (e.g. Remains or body part). This is also similar to another recent request: https://github.com/nvs-vocabs/OBISVocabs/issues/25

There's a couple of option for how we could handle this in the P01 vocabulary: Option 1: For each occurrence you would have an eMOF with one of these measurement types (some of these may not exist yet, but we can create new terms in the P01 vocabulary if needed): Count of biological entity specified elsewhere Count of biological entity specified elsewhere [Subcomponent: shell] Count of biological entity specified elsewhere [Subcomponent: exuvia]

Then each occurrence also have an eMOF with meaurementTypeID: https://vocab.nerc.ac.uk/collection/P01/current/SDBIOL11/ measurementType: Categorisation (as living or non-living) of biological entity specified elsewhere measurementValue and measurementValueID could be from S13 (dead/live)

Option 2: Each occurrence would have these three measurement types linked in the eMOF table: measurementType: Count of biological entity specified elsewhere

measurementType: Categorisation (as living or non-living) of biological entity specified elsewhere Measurement value: from S13 vocabulary e.g. dead/live

measurementType: Subcomponent/body part of biological entity specified elsewhere (we would need to create a new term for this, and this has actually already been requested: https://github.com/nvs-vocabs/OBISVocabs/issues/25) Measurement value: from S12 vocabulary e.g. shell/exuvia

Which option is preferable for VLIZ/OBIS?

Many thanks, Roseanna

JoBeja commented 2 weeks ago

Hey Roseanna, I've made my colleagues aware and they will provide input when they're available (some are currently on leave).

LynnDelgat commented 1 week ago

Hi, my preference goes towards the second option. Having a separate body part term seems more flexible, as this could perhaps be useful to combine with other parameters than count as well. Plus there could be quite a lot of possible values for subcomponent/ body part, so that would result in a lot of new terms in the case of option 1.

roswri commented 1 day ago

Thanks Lynn,

So the approach would be that we would create a new P01 for 'Identifier of subcomponent of biological entity specified elsewhere' and the measurement value would be recommended to be taken from the S12 vocabulary.

One slight issue is that we have 19 existing 'biological entity specified elsewhere' P01 terms which have subcomponents within the term: https://vocab.nerc.ac.uk/search_nvs/P01/?searchstr=biological%20entity%20specified%20elsewhere%25subcomponent&options=identifier,preflabel,altlabel,status_accepted&rbaddfilter=inc&searchstr2=

We would ideally like our approach to be consistent, so I think we need to discuss this further with OBIS.