Taxonomic Swap 133177
(Guardado el 30/10/2023)
|
|
Reemplazado con |
|
Comentarios
@loarie this was already committed, we might need a hand with this.
@silversea_starsong Observations that were identified to E. o. occidentalis and E. o. hellmayri are transferred appropriately, but to move everything from Cordilleran to E. d. occidentalis? There's a reason this lump happened in the first, because the two are literally indistinguishable. Listen to any "Cordilleran" from the Northern/Middle Rockies, they are not Cordilleran. You don't get real Cordilleran until at least Colorado. But the flycatchers from Arizona are definitely not the same birds from Montana. But regardless of opinion, unless you're hearing the bird and even then, it's highly unreliable, you aren't going to be able to identify 95% of these birds to subspecies.
It's not really our decision to move things on behalf of others this way. If they are labelled as occidentalis they should go to ssp. occidentalis, and if the ID is then disputed, it can be dealt with on those observations. I totally understand your point, but this is not really how the taxon changes are handled. By that argument though, it acknowledges too that a very moderate amount of occidentalis and difficilis are now lumped when they have "pretty good" merit to their ID.
But, well, maybe it is the right call after all. I just don't think it is how iNat is "supposed" to handle this.
I am not a curator (and I usually hate subspecies!), but I agree that the way this split was set up doesn't seem right. Under this mindset, why bother keeping the subspecies in the taxonomy at all? What about the 700 observations that have audio and could conceivably have been separated? And now this deprecates identifications happening pre-split (which all just got nuked) vs post-split (which will generally stand unless individually curated), despite both usually being based off the same information (call and range).
I'm shocked this one is getting so such backlash for such an ambiguous species. Let me see if I can explain this a little better.
"I just don't think it is how iNat is "supposed" to handle this."
I don't want to put words in the staff's mouth, but Loarie was the one committed the Crested Caracara lump. Where was the backlash for that? 14,000 Caracara planus cheriway observations all bumped to species level, when we could've just handed it over to subspecies rank.
"What about the 700 observations that have audio and could conceivably have been separated?"
That's precisely why they were lumped in the first place. We know that call is not diagnostic, because everything in the North/Middle Rockies sound intermediate, and we know through backplays in Hopping (2022) that these birds can change their calls. There are a number of recordings of Cordilleran switching to the Pacific-slope pseet call. Heck, the two are so indistinguishable, that even banders would even try. So, you can literally have the bird in your hands, and you still couldn't be certain.
Just to make it clear here, the argument is not that these two taxa are indistinguishable, or when and where they should be identified. It's that taxonomy changes are solely the process of transferring names, not as a direct solution to address misidentification. So, occidentalis is now a subspecies and should be transferred that way. You are quite right that the ID is difficult (hah, difficilis) and that most of these probably shouldn't be identified, but that's outside of the scope of a taxon swap. The taxon swap is not designed to be a solution to "fix" an ID issue in the case of a 1-to-1 transfer, even if it can do so in a secondary way.
So this is not really "backlash" as much as it's just not how taxon changes should be done, which is why I said it's not what iNat is "supposed" to "handle this".
I can't speak for Scott on the caracara split, but I presume that was an oversight.
I understand your point of view, but I'm not sure you see the fallout if I had done it the other way. The Cordilleran Flycatcher is a polytypic group, and iNat's system does not allow a good way to report to polytypic subspecies groups. If I have transferred all Cordilleran into E. d. occidentalis, then we'd have well over 1,000 obs from Mexico that would be applied to the wrong subspecies, because they're hellmayri, not occidentalis. That's why you don't see reports of Oregon Juncos on iNat, because we don't have an option for a polytypic group. I know it's an Oregon Junco, but how am I'm supposed to know which Oregon it is since it's a group of six nearly indistinguishable subspecies. Same thing goes for Fox Sparrows. I can know the bird I'm looking at is a Sooty Fox-Sparrow, but if I'm going to report it as such on iNat, I'd have to prove beyond a shadow of a doubt that's it sinuosa or something like that, and that is nearly impossible.
Right, yeah no I definitely see the virtue to doing it this way. And I'm actually more on the fence than my comments might suggest. But for me to firmly agree, I'd have to know more precisely what % of the occidentalis have virtue as that (sub)taxon to decide if the means justify skirting the usual taxon swap rules. I'm sure it's not very high, but there's plenty of breeding ground records that would fit.
What you're describing makes it sound like this was already an issue, with "well over 1,000" obs from Mexico that were already misapplied as Cordilleran? Or am I misreading.
@silversea_starsong No, hellmayri is a subspecies of Cordilleran from Mexico. Since Cordilleran in pre-split was a polytypic species, there's no way to indiscriminately transfer observations to occidentalis. Sure, we could've atlased all Cordilleran in accords to their breeding distributions, but occidentalis winters side by side with hellmayri. And most folks were identifying Cordilleran Flycatchers, not the hellmayri subspecies of the Cordilleran Flycatcher.
This would have been good information to give earlier, since that feels like a pretty important facet to the decision over how this was done.
Unfortunately I do not think there is a way to atlas based on the time of observation (since no doubt the migration birds are not as safe to call).
@birdwhisperer this should have been transferred to subspecies, not the full species, as this loses all the inherent data by merging it all into one unspecified heading.