Flag for Taxon: foamflowers (Genus Tiarella)
| Flagger | Content Author | Content | Reason | Flag Created | Resolved by | Resolution |
|---|---|---|---|---|---|---|
| trscavo | foamflowers (Genus Tiarella) |
Migrate from Tiarella cordifolia to Tiarella stolonifera in Québec |
Nov. 12, 2022 16:36:53 +0000 | rynxs |
swapped |
Comments
So in Quebec change Tiarella cordifolia to Tiarella stolonifera? I can work on it.
Is there any way to swap these ID automatically? Otherwise, this will take a while, especially for T. cordifolia with 4-5 IDs.
If one was migrating records, it would make sense to extend that to all areas where there is no issue of range overlap.
@seanblaney I chose Québec as a starting point since I thought that might be easier. We can start there and see how it goes, or we can try to tackle the entire problem at once.
There are 17 provinces and states with one taxon of Tiarella (just like Québec):
New Brunswick (59), Nova Scotia (17), Ontario (1369), Québec (350); Connecticut (62), Maine (112), Massachusetts (424), Michigan (160), New Hampshire (392), New Jersey (45), New York (1230), Ohio (658), Pennsylvania (744), Rhode Island (2), Vermont (1444), West Virginia (215), Wisconsin (17)
The number in parentheses is the number of observations that contain an ID of Tiarella cordifolia. Altogether there are 7299 observations with one or more IDs that can be swapped from Tiarella cordifolia to Tiarella stolonifera.
Tackling the entire problem at once seems the better option, otherwise many observations may be lost in the limbo of Tiarella sp. This would be a huge time saver too.
We could also automatically downgrade to Tiarella sp. all observations of Tiarella from the regions of overlap with other species, this would help start things from a blank slate instead of having to deal with contradicting IDs
Or we can create a "Tiarella cordifolia complex" where we could put all these observations
@elacroix-carignan I have a list of states (no provinces) for which a downgrade to genus Tiarella would be justified but I think that would be better handled in a separate flag. Regarding a "Tiarella cordifolia complex", I am not aware of any justification for that in the literature. Do you have a reference?
@maryah would you rather focus on Québec or migrate all 17 provinces and states at once?
I’ll start with Quebec.
Thanks @maryah
@maryah going back to a question posed by @elacroix-carignan earlier in the thread, are you migrating these IDs manually or automatically?
@maryah appears to be "migrating" them by manually re-identifying them.
@trscavo what splits need to be put together? You don't need to bother with manual re-identification if you can just explain which taxa need to be split. If I'm understanding this right, T. cordifolia needs to be split into T. stolonifera and T. cordifolia. How do T. austrina, T. wherryi, and T. nautila fit in?
Tiarella doesn't contain that many species, so re-identifying genus-level observations probably won't be all that bad if we have multiple people tackling it. Do you have a publicly available key?
Finished Quebec. I’ll help some more tomorrow confirming observations in other areas.
Thanks @maryah that's very helpful!
@rynxs both T. cordifolia and T. wherryi have been split. T. stolonifera and T. austrina were split from T. cordifolia. T. nautila was split from T. wherryi and T. cordifolia. If it's easier to think of T. nautila as being split from T. wherryi alone, then we should probably make that simplifying assumption.
The 17 provinces and states listed above represent a potential optimization. Since there is only one taxon in each of those 17 provinces and states, I consider the split to be little more than a name change there. I hope you do too.
It took about 2 hours. Now if others can confirm them. Some will need two more for research grade. If there is someplace like Quebec with no other choice let me know. They are difficult to tell apart. But like I said I will look for other ids to confirm.
If you can do it automatically, all the better!
So, in essence:
Tiarella cordifolia s.l. split into T. austrina, T. cordifolia s.s., T. nautila, T. stolonifera
Tiarella wherryi s.l. split into T. nautila, T. wherryi s.s.
Considering the range overlap, I think manually re-identifying T. nautila from T. wherryi would be easier than a split. I will work on atlasing these species so that they can be split. There doesn't seem to be much point in adding subgenera at the moment, so T. cordifolia s.l. observations will have to be re-identified from genus level where ranges overlap.
@rynxs take a look at this Proposal to revise Tiarella on iNat
@trscavo do you know how splits work on iNat? Atlased split taxa are automatically brought to the correct taxon if it is the only taxon for the state. I cannot swap individual states, but I have set up a split: https://www.inaturalist.org/taxon_changes/116684
At the moment, T. cordifolia needs to be re-atlased.
Sorry @rynxs I don't know anything about iNat curation
In essence, an atlas is a range map for a taxon. Adding regions means that the species occurs there, which is factored into splits. If one product taxon of a split occurs in a region, IDs on observations of the split taxon will be moved to the product within that area. If two product taxa in a split share the same region, the taxon being split will instead be moved to the most specific common taxon between those two (in this case the genus Tiarella).
I have removed a significant number of regions from T. cordifolia's atlas. Introduced populations in Minnesota, Illinois, Indiana, and Europe will need to be reviewed, as well as those not on the following list (specifically Delaware). All observations not placed in a region on this list will be moved to genus:
Alabama
Connecticut
Georgia
Kentucky
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Vermont
Virginia
West Virginia
Wisconsin
Norway
You can check atlases by reviewing the maps on a taxon's "About" page. Zoom in to see the atlased places.
https://www.inaturalist.org/taxa/1426513-Tiarella-austrina
https://www.inaturalist.org/taxa/53789-Tiarella-cordifolia
https://www.inaturalist.org/taxa/1426514-Tiarella-nautila
https://www.inaturalist.org/taxa/1351707-Tiarella-stolonifera
Thank you @rynxs for this little crash course on how atlasing works! I figured there would be a way on Inaturalist to reassign species based on the region of occurrence, that will save some of us a lot of time!
I finally found the atlases. I'll comment on each one separately.
Tiarella nautila https://www.inaturalist.org/atlases/63160
This looks right (Georgia, North Carolina, Tennessee)
Tiarella austrina https://www.inaturalist.org/atlases/63159
This atlas appears to be blank? In any case, T. austrina occurs in Alabama, Georgia, North Carolina, South Carolina, Tennessee.
Tiarella stolonifera https://www.inaturalist.org/atlases/63161
This atlas is mostly correct (but it's kinda hard to tell from a map). T. stolonifera occurs in Connecticut, Kentucky, Maine, Maryland, Massachusetts, Michigan, New Brunswick, New Hampshire, New Jersey, New York, North Carolina, Nova Scotia, Ohio, Ontario, Pennsylvania, Québec, Rhode Island, Tennessee, Vermont, Virginia, West Virginia, Wisconsin. Note that Minnesota and Norway are not included in the list. Although POWO includes them, there is no evidence for doing so (as far as I know). Since those two populations were introduced, they could have been sourced from anywhere.
I can't find an atlas for Tiarella wherryi. Is there one?
An atlas for each species of Tiarella is given in wikipedia. I created those lists myself so I know they're correct :-)
@rynxs the list of states that you typed above is correct except for Minnesota and Norway. As I mentioned in my previous comment, I wouldn't include those but it's your call.
Did you intentionally omit the list of Canadian provinces from your list, or was that an oversight? In any case, the relevant list of provinces from Canada is: New Brunswick, Nova Scotia, Ontario, Québec.
No, it wasn't intentional. Thanks for reminding me.
Sometimes atlases glitch and aren't visible to some users, but they are all atlased according to POWO's range maps.
I excluded Minnesota and Norway from T. stolonifera's atlas.
If no problems occur, I'll split by next week.
I can see the atlas for Tiarella austrina now (linked above). Also, the atlas for Tiarella cordifolia looks good: https://www.inaturalist.org/atlases/3407
Overall, the proposed taxon split looks good: https://www.inaturalist.org/taxon_changes/116684
@rynxs what about Tiarella wherryi?
As I said above, I think (considering the small quantity of T. wherryi observations) a split on T. wherryi would do more damage than good. With only 96 observations IDed as T. wherryi it might take a few hours of identifier effort to correct, where a split would reduce every one of those observations back to genus level, making them harder to find.
Here's a link to identify every Needs ID/RG observation ever identified as T. wherryi: https://www.inaturalist.org/observations/identify?quality_grade=needs_id%2Cresearch&ident_taxon_id=244487
@rynxs I think you're right, and actually, I'm glad to hear you say that. The observations of T. wherryi that we have now were hard-won, and some people might be unhappy to lose them.
@rynxs maybe next week?
Committed, since 7 days have passed. This change will take a while to go through.
Thanks @rynxs! I very much appreciate your assistance and expertise!
I do not concur with this change. I believe it is premature. The split has not yet been accepted by Flora of North America. It has not yet been accepted by the Integrated Taxonomic Information System. It has not yet been accepted by the New York Flora Atlas. Until these organizations accept the proposed change, a consensus does not exist. I strongly suggest that iNaturalist wait until those organizations accept the proposal before making this change.
http://floranorthamerica.org/Tiarella_cordifolia
https://www.itis.gov/servlet/SingleRpt/SingleRpt?search_topic=TSN&search_value=24529#null
@charlie, @susanelliott
FYI
@ellenjones6 have you read Nesom's paper?
http://www.phytoneuron.net/2021Phytoneuron/31PhytoN-Tiarella.pdf
@rynxs, glad you link to the paper. For anyone curious, there's a map of the split species on page 10 of the paper that's worth looking at to understand how the new species are split geographically. I'm not sure what happens in overlapping areas or ones not shown.
@cgbb2004 i cant be involved with this any more, i dont have any more emotional capacity to deal with the profession and personal attacks i get for disagreeing on inat taxonomic policy.
The taxon split renamed many casual - cultivated - observations of T. cordifolia to stolonifera. Two examples:
https://www.inaturalist.org/observations/23059109#identification-c420125d-301c-4178-b69a-920da97ddbb7
https://www.inaturalist.org/observations/23198021#identification-a5b9bbc5-c033-4e20-8191-56bf64e5760c
T. cordifolia (in its original sense) is widely planted as an ornamental in Eastern North American gardens. Since the origins of planted specimens can't be determined from photographs, it's inappropriate to reassign them from a range-based taxon split.
@xris, are you thinking the captive/cultivated ones should be up-leveoed to genus for the split?
@vreinkymov: Yes. I noticed this from observations in New York City:
Most, if not all, the Tiarella observations here will be cultivated, and should be flagged as "not wild". I'm correcting those as I come across them.
Per New York Flora Atlas, there are no records of Tiarella cordifolia occuring spontaneously in NYC.
I've also been able to rule out stolonifera for several observations that were moved with the split by the presence of leaves on the stems shown in the photos.
@xris @vreinkymov the split was executed on November 20, 2022. Some observations were converted that shouldn't have been converted. I fixed most of them. I'm currently in the process of reviewing the casual observations, downgrading them one-by-one if needed. I have about 30 pages left to go. If you want to help, please do.
@trscavo Yes, it was your reassignment on those two observations I noted above that alerted me to the change!
Right now, I'm going through all the NYC RG Observations of Tiarella to correct them to Casual, and upping them to the genus. As I go, I'm checking them against stolonifera to see if that can be ruled out from the photo. (Most are only of the flowers and don't show the characteristics needed for species-level identification.)
POWO recognizes five taxa of Tiarella in eastern North America, but there is only one taxon over a very large region. For example, consider Québec, the most active iNat user of Tiarella stolonifera at this time. There is only one taxon of Tiarella in Québec. Some people call it Tiarella cordifolia while other people call it Tiarella stolonifera:
Number of observations that contain an ID of Tiarella cordifolia: 350
Number of observations that contain an ID of Tiarella stolonifera: 53
Of those 403 observations, 14 of them contain both IDs.
We should swap Tiarella cordifolia with Tiarella stolonifera in all 350 observations so that Québec ends up with 389 observations containing an ID of Tiarella stolonifera (and no Tiarella cordifolia).
Can this be done?