Splitting sympatric taxa as cleanly as possible

Curators
, ,
1

Hello all—I am working out how to split two heavily sympatric plant taxa with as little disruption as possible. I wanted to ask for advice as iNaturalist’s existing documentation (here, here) for splits deals primarily with hypotheticals where the two output taxa are allopatric.

The species are Valeriana sitchensis and Valeriana scouleri (see flag). Valerian scouleri currently exists in iNat’s taxonomy as V. sitchensis scouleri but has recently been re-elevated to species rank. The range of V. sitchensis sensu stricto and V. scouleri overlaps in southern British Columbia, Washington, Oregon, and northern California, where they occupy different, but adjacent, elevation niches that cannot be atlased. (V. scouleri grows at lower elevations; V. sitchensis at higher elevations.)

First, I created a V. sitchensis complex and made it the parent of V. sitchensis and the inactive V. scouleri so that any conflicting IDs caught up in the swap would be caught there instead of elevated to genus.

Next, I atlassed both V. scouleri (atlas) and V. sitchensis (atlas):

V sitchensis atlas:

V sitchensis atlas
V sitchensis atlas1546×1580 634 KB

V. scouleri atlas:

V scouleri atlas
V scouleri atlas1398×1602 604 KB

Then, following iNat’s documentation—

the input taxon has two active children (subspecies) that must first be handled with two separate taxon swaps into the respective species…before the split can be committed

—I created swaps to swap V. sitchensis scouleri into V. scouleri and V. sitchensis sitchensis into V. sitchensis. My intention was to only commit the split of V. sitchensis into V. sitchensis and V. scouleri after the swaps had been committed, i.e.

  1. Swap V. sitchensis scouleri into V. scouleri
  2. Swap V. sitchensis sitchensis into V. sitchensis
  3. Split V. sitchensis into V. sitchensis and V. scouleri

This follow’s iNat’s documentation, but here’s my concern: Because this swaps V. sitchensis sitchensis into V. sitchensis before the split, my understanding is that this will destroy the detail in IDs for V. sitchensis sitchensis where the two output taxa’s atlases overlap. Because V. sitchensis sitchensis will be subsumed into V. sitchensis first, when the split occurs, all those IDs of V. sitchensis sensu stricto in the area of overlap will be elevated to the level of complex—V. sitchensis sensu lato.

Would it not make more sense to instead follow the steps in this order?

  1. Swap V. sitchensis scouleri into V. scouleri
  2. Split V. sitchensis into V. sitchensis and V. scouleri
  3. Swap V. sitchensis sitchensis into V. sitchensis

Any advice would be very much appreciated! Thank you!

2

I think I agree with your proposed sequence (splitting V. sitchensis before swapping susbp. sitchensis into its parent). That ensures that the information contained in existing IDs of subsp. sitchensis is preserved.

Because taxon splits can now use the source taxon as one of the outputs, this seems like it should work fine. Definitely get some other opinions though, as I’m no-one’s expert on taxon changes!

I would consider whether you can fine-tune either of the atlases at county level, although I appreciate that the sympatry and elevational distinction don’t give you much to work with.

As for the complex… I see that it’s a good solution to prevent current V. sitchensis IDs getting bumped up to genus. Does it also reflect a phylogenetic and morphological relationship? Is it something that should persist long term?

3

I think what you propose should work. Just make sure that you wait until your first swap is fully processed before splitting.

4

Thank you both! Appreciate your help!

Yes, the V. sitchensis complex is phylogenetically valid—the two are sister species and very similar morphologically, so I think it serves a useful purpose long-term.

5

Is it recognized in the literature as a complex outside of iNat? This is one of the requirements for adding a species complex to iNat:

https://www.inaturalist.org/pages/curator+guide#complexes

I don’t see an issue with creating a phylogenetically valid complex as a short-term placeholder taxon to facilitate a better taxon split, but if the complex doesn’t meet iNat’s criteria, it shouldn’t be maintained after the split is completed.

6

Good question! It is not, and for that reason (among other more pressing taxonomic ones) raised by @ajwright in the discussion on the flag here, I will actually be removing it before moving forward with the split: https://www.inaturalist.org/flags/839760. As a result, the split will elevate roughly 12,000 IDs to genus level, but we will be following iNat’s curator guidelines to the letter.

I’ll note that I’ve seen a lot of species complexes created for the purpose of enabling a cleaner split in the past without justification in the literature. (I’ve even got pushback when I raised the issue on flags before.) I am grateful to you all for maintaining strict standards here.

Appreciate the help, everyone!