When some subset of species are moved from Examplegenus to Newgenus this implicitly changes the meaning of all previous IDs of “Examplegenus sp.” (that is, IDs at the genus-level) from a broad sense to a narrower sense, now excluding the moved species.
How is this best handled?
If I understand you correctly, I think you can see a similar case with herring gulls where the recent split left many old observations with IDs of “European Herring Gull” which don’t match the identifiers’ original intentions of L. argentatus sensu lato (e.g. obs 1, obs 2). At the time, that ID had a broader meaning.
Explanation here:
It’s unclear to me if that’s a recommended practice to reduce server load or if there’s another reason it’s being done this way.
Depending on the number involved, either reviewing all affected observations or splitting the old genus (with an atlas when possible), which will bump IDs to the next lowest level.
I understand the related problem/policy of iNat taxa not being explicitly wedded to a particular sense, which is the subject of the linked-to discussion. I’m asking about best curator practices for dealing with one of the consequences of this problem/policy – what to do when previous IDs at a higher-level have implicitly changed meaning.
The herring gull example shows the system working as intended. With the exception of taxon change opt-outs, the system handles Genus species (s.l.) –> Genus new-segregate-species + Genus species (s.s.) cases pretty well. In this case, only “Genus species” has had its meaning changed. “Genus” itself, has not changed meaning.
I am concerned with cases where daughter taxa have been moved in a way that changes the meaning of the mother taxon. For example:
Genus species1, Genus species2, Genus species3, Genus species4 –> Newgenus species1, Newgenus species2, Genus species3, Genus species4.
After this change, previous instances of “Genus sp.” IDs given at the genus level may be **outright incorrect** because they may have referred only to G. species1 or G. species2, which are now in Newgenus.
https://forum.inaturalist.org/t/new-tool-for-analyzing-unintended-disagreements-caused-by-thinning-parents/31970 has some good info
Thank you! I tried to search for this, but I couldn’t conjure the right keywords.
It looks like the rule-of-thumb answer to my question is:
if thinning a parent results in more than 10 unintended disagreements you should split the parent afterwards to replace existing IDs of the parent with IDs that don’t disagree.