Hi everyone! We (Nadiia @skobel2015 and me) are preparing a dataset with species occurrences from the vegetation database, and we couldn’t find a good solution for publishing species aggregates and species sensu lato taxa. For example, we have in our dataset: Festuca valesiaca agg. and Galium verum s.l. And these are not the same taxa as Festuca valesiaca and Galium verum.
I see a list for some of the aggregates here - Search , but without those which we need.
I think you should be able to publish occurrences with the aggregate name in the scientificName field, and aggregate in the taxonRank field as well as providing the associated higher taxonomy. This should limit the possible misinterpretations.
You could also consider publishing a checklist with the aggregate names that you will use in your occurrence data (note that occurrences can be an extension of a Taxon core).
@markus what do you think? Does that seem ok to you?
Publishing occurrences with those names is just fine and GBIF can deal with them.
But we do not differ between the taxa in our backbone yet. Unfortunately there will only be a single Festuca valesiaca and Galium verum, no matter if seno lato stricto, auct amer or what else. We know this is limiting, but at this stage we cannot handle multiple taxa with the same name.
So do you think we should publish these occurrences e.g. with scientificName = Festuca valesiaca and taxonRank = aggregate? I understand that some people assume these aggregate names to be not very useful. But in our type of data (sampling of vegetation), we sometimes have these types of identifications, and we want to preserve them somehow to avoid misunderstanding for narrow specialists (mismatch with Festuca valesiaca s.str.).
I also think that we can put the aggregate name into verbatimIdentification field - a possible option to have it preserved.
I think you can use scientificName = Festuca valesiaca agg. and taxonRank = aggregate
When you use the scientificName field, users have access to both the name that you provided and the interpretation by GBIF (which, in this example, would be Festuca valesiaca Schleich. ex Gaudin). So the information will be preserved, users will know that the original name provided was Festuca valesiaca agg.
Using the verbatimIdentification is also a good idea. There is nothing preventing you from doing both.