Flag for Taxon: Jack-in-the-Pulpit (Arisaema triphyllum)
| Flagger | Content Author | Content | Reason | Flag Created | Resolved by | Resolution |
|---|---|---|---|---|---|---|
| bouteloua | Jack-in-the-Pulpit (Arisaema triphyllum) |
should be split if new taxa are being carved off it |
Jan. 31, 2023 04:10:31 +0000 | bouteloua |
maverick IDs to remain |
Comments
@loarie my identifications of Arisaema triphyllum sensu lato are being disagreed with individually rather than handled through a taxon change and bumped to a coarser taxon based on new species being carved off. Is there a change to policy I am out of the loop on?
Its a judgement call on when to do a retroactive split to automatically roll back IDs or sort it out via IDs.
The downside of the retroactive split is that all obs ID'd at Arisaema triphyllum will get rolled back if the outputs of the split are sympatric. The downside of manually sorting out IDs is that it can take alot of new IDs to in some cases roll back and moreso roll forward IDs to taxa when there are disagreeing IDs left in place.
The rule of thumb used in the tutorials is if there's more than 10 unintended disagreements you should split
https://inaturalist.freshdesk.com/support/solutions/articles/151000015334-overview-on-resolving-taxon-flags
but for these huge taxa with many tens of thousands of IDs that criteria might be met with only a very tiny tiny fraction of unintended disagreements.
What's your preference?
I think it's unfair, confusing, misleading, and frustrating to impose unintended disagreements on other users, so my preference is the status quo.
Can you construct a draft retroactive split/atlases or explain the structure here so others can make one?
I'm not deeply familiar with the taxonomy and don't plan to curate this myself (I haven't been doing much curating lately), I and others have just been receiving notifications of disagreeing IDs and am raising this to attention as a curation need.
I'm going to make it clear that I don't want to deal with the aftermath of this if you choose to go through with it, although I will continue to help as an identifier. The only way to perform a split here is to take every single Arisaema triphyllum observation and elevate it to complex level, which then requires manual review of 50,000 observations anyway.
How do I block my IDs from being elevated to complex, since they're all for Arisaema triphyllum sensu stricto?
There's no way to prevent your IDs from being update by a taxon change unless you opt out of taxon changes entirely (which I don't recommend).
The status quo for retroactive split has always been some potentially helpful rule of thumbs behind a judgement call based on a set of tradeoffs:
the confusing and frustration imposed by unintended disagreements that bouteloua mentioned
and
the confusing and frustration imposed by observations getting coarsened that rynxs mentioned
My preference is that the decision is made based on a curator stepping forward to ringlead. ie if someone is willing to review obs, add new identifications and comments to try to work through unintended disagreements and the confusion caused by the addition of the new taxa than thats probably easiest
likewise if someone is willing to construct the split and review obs, add new identifications and comments to try to work through the frustration and confusion caused by the coarsening/loss of taxonomic precision from the split than thats probably easiest
It sounds like rynxs has looked into these tradeoffs and made a judgement call and is willing to do the work to minimize the impact of unintendend disagreements so that seems like a good path forward to me
If anyone has recommendations on how to better balance these tradeoffs and engage people to do the needed curation work to execute them I'd definitely appreciate feedback
As someone not deeply familiar with the taxa involved here, I'd personally probably benefit most from a journal post describing the different taxa involved and how they're different and how to distinguish them by characters and or geography (something like this) regardless of whether I'm working through confusion about unintended disagreements or about why my obs was coarsened. If anyone has time to draft something like that, I think it would be a very useful link to include in comments to help with confusion
As your "preference" above is a change to policy please reflect that in the guidance.
I reviewed https://help.inaturalist.org/support/solutions/articles/151000015334-overview-on-resolving-taxon-flagsand don't see anything written there thats not consistent with the current situation where we lack precise rules. If you see any documentation that needs to be updated please let me know.
Likewise if you can articulate a precise rule that you think would reduce confusion, I'd be happy to start some discussions with various stakeholders/users and compile responses
@bouteloua @loarie I've just found this out in the past few days, but apparently there will likely need to be a split anyway in the future. An undescribed entity currently named "Arisaema species 2" is set to appear in the April edition of Weakley's Flora of the Southeastern US. This entity probably encompasses most of the observations on iNaturalist outside of the Midwest states. Obviously a split can't happen until the element is described (and POWO accepts it). I made an observation field for it ("Arisaema triphyllum complex") and the undescribed element in Arisaema dracontium as well ("Arisaema species 1"). I may have also found another undescribed species using iNat observations, but I'll have to ask around to see what others think.
I went ahead and made a very rudimentary map using those observation fields to get a feel for where the populations are. Interestingly, the distribution of Arisaema triphyllum sensu stricto lines up fairly well with Polygonatum biflorum, and the distribution of "Arisaema species 2" appears to line up with Polygonatum pubescens, aside from the southwest population groups. I wonder if they have similar evolutionary histories?
I am manually reviewing observations. This will take longer, but there's no need for an automated split when >95% of observations are of A. triphyllum sensu stricto, and there aren't any clear geographic boundaries to base atlases on.