Closed rukayaj closed 1 year ago
I was hoping that we could keep each entry as one record, even though there are several taxa, and mention them by the highest common level, e.g. "Bryozoa" for sample 667. There are several taxa in many of the samples, maybe in most of them, but its just for some that I have made a note about which they are. Generally we haven't recorded (or do not know) what the taxa are. Is this feasible, within GBIFs framework, to register bulks like this, with only the highest common taxonomic level mentioned? We can of course delete this kind of information from the verbatimIdentification-field. It is more "additional arbitrary information", not recorded in a systematic fashion, and I guess there is no dwc-field for that.
I agree entirely that dwc:MaterialSample entries (including notably the environment samples for eDNA) would often be bulk samples composed of multiple taxa, or even no taxa at all! I hope the new GBIF data model under development and enabling the move away from the Occurrence model straight-jacket will enable better representation of such bulk samples -- soonish!
And if a bulk sample includes a mixture of organisms from multiple kingdoms, I think that the super-class biota
is completely legitimate to use
https://www.catalogueoflife.org/data/taxon/5T6MX
https://www.gbif.org/species/170809336
Thanks dag! Ok that sounds like a good move then, we keep 1 row per bulk sample, as you've got it currently, and use biota or whatever higher taxonomic level.
@gunnhilm I've made a dataset on the IPT for this - https://ipt.gbif.no/manage/resource.do?r=uio-nhm-bulk, I'll send you user credentials in an email.
The collection consists mostly of bulk-samples - so this makes some of the datafields more challenging - e.g. when there are more than one taxa in one sample. The collection is registered in GRSciColl; "Collection for bulk- and project material".
The database for the collection is so far only an excel-sheet, e.g: