search

nvs-vocabs / EMODnetChemVocabs

A repository for the management of issues related to vocabulary concepts and collections used by the EMODnet Chemistry project.
1 stars 1 forks source link

Could we bypass P02 when linking P36 to P01 and use S27 instead? #42

Open gwemon opened 1 year ago

gwemon commented 1 year ago

See https://github.com/nvs-vocabs/EMODnetChemVocabs/issues/34 for background to the issues and difficulties in aligning P36 and P02 bearing in mind that a P01 can only be linked to one and only one P02. This results in the harvest of datasets that are irrelevant to the P36 categories because the P02 code relates to a chemical grouping that is broader than the P36 one. To illustrate this I have updated the EMODnet_chem thesaurus so that it now displays the chemical substances of the P01 codes that are linked to the P02s associated with the P36 concept. By navigating this scheme you will quickly notice some anomalies: for example, under Antifoulants, we have some chemical substances that are not related to antifoulants, and the reason for this is because the P02 category is not specific to antifoulant chemicals but is based instead on organometallic and organometalloid species grouping. Under Organic matter we have inorganic carbon because a P01 code related to inorganic carbon belongs to the P02 Carbon concentration in sediments

The only way around this is either to split the P02 categories and remap existing P01 (heavy work that would also impact vocab maintenance workload) or bypass P02 and obtain the P01 from the mapping between P36 and S27.

@SLBlakeman FYI

MeganAnneFrench commented 1 year ago

Hi Gwen, Marina is on holiday this week, back next week, and then on holiday again. I will remind her about your query, but if you don't hear back from her until September, this is the reason! Megan

gwemon commented 1 year ago

I will share with EMODnet chemistry technical group too. Thanks @MeganAnneFrench.