Heads up: Some or all of the identifications affected by this split may have been replaced with identifications of Pachycephala. This happens when we can't automatically assign an identification to one of the output taxa. Review identifications of Pachycephala pectoralis 506043

Taxonomic Split 99303 (Committed on 2022-04-11)

Clements Checklist v2021 (Citation) | split#62 - Pachycephala pectoralis
| Committed by loarie on April 11, 2022
split into

Comments

Joseph et al. 2021

Posted by jwidness 9 months ago (Flag)

Currently 26 obs are getting bumped to genus, but unclear if that's too conservative or not conservative enough. The above map from Joseph et al. shows a fairly large gap between samples, but they do provide a dividing line. The same paper also shows a map that's based only on plumage, which suggests the division should maybe be further east?

@thebeachcomber @davidsando @ratite you've identified birds in this area, can you say anything about where the split should be?

Posted by jwidness 9 months ago (Flag)

So is there a zone/threshold at the junction of the two areas in which records go back to genus?

Posted by thebeachcomber 9 months ago (Flag)

The way the atlases are set right now, these 26 observations will have their IDs raised to genus: https://www.inaturalist.org/observations?id=96763793,94197866,92896427,92542525,90129751,90129639,82663818,65980425,58491827,54902694,52320523,44955765,44956343,44341779,42817005,41892155,41151514,40944048,40732500,40221698,40218570,38171113,36371127,35515991,35401526,34101334&place_id=any&subview=table&verifiable=any

The question is whether that list can be narrowed down (e.g. because you can look at the photos and can tell which way to split), or whether it should be broadened (e.g. because the maps from Joseph 2021 aren't precise enough).

Posted by jwidness 9 months ago (Flag)

I can't truthfully say I'd be able to separate the two here looking at the plumage in those pics, so I would personally be in favour of broadening it a bit

Posted by thebeachcomber 9 months ago (Flag)

I'd agree with keeping it a little bit broad like it is, just as the plumage features can be a bit tricky from looking at the photos. Unfortunately I wouldn't be able to confidently separate them either

Posted by ratite 9 months ago (Flag)

here's the current summary -- fuliginosa atlas and range map in green, pectoralis atlas and range map in orange, red points will get bumped to genus, blue points on the left (e.g. around Adelaide) go to fuliginosa, blue points on the right (e.g. around Melbourne) stay pectoralis

Posted by jwidness 9 months ago (Flag)

looks good to me; unfortunately there won't be a notification for users when theirs get bumped to genus right? I'll go through each of those 60 odd and let people know if that's the case

Posted by thebeachcomber 9 months ago (Flag)

The IDs won't get bumped if they've opted out of having IDs automatically reassigned. For those that haven't opted out, there should be a dashboard notification, but I don't think it shows up in the header notifications.

Posted by jwidness 9 months ago (Flag)

any timeline on this swap?

Posted by thebeachcomber 6 months ago (Flag)

Also worth noting there seems to be a population of Western Whistler in the whipstick north of Bendigo which won't be picked up by this filter swap.

Posted by scarletmyzomela 6 months ago (Flag)

@jwidness @loarie Something has gone wrong with this split as there are observations that there should be no doubt about that have been reverted to genus
e.g. https://www.inaturalist.org/observations/38659843#activity_identification_bf54de81-0033-4642-ad00-999f648724a7

Posted by rfoster 3 months ago (Flag)

Everything in region Adelaide Hills-Bal, SA, AU seems to have been affected. It seems it wasn't properly atlased when the split was committed. Not sure if restricted to this place or is a wider problem.

Posted by rfoster 3 months ago (Flag)

So we're on the same page, nothing unexpected happened here. The atlases were configured to replace nearly all (all but 182 out of 7344) existing IDs to one of the outputs with just 182 being replaced by the genus. The idea is that anything coarsened by those 182 IDs replaced with genus can be refined manually - here's a link to the obs in question if it will help people spend some time ID'ing them

Posted by loarie 3 months ago (Flag)

That doesn't explain why observations in the Adelaide hills got "coarsened" unnecessarily as there is no doubt that they are P. fuliginosa based on locality. The affected records are not included in the link.

Posted by rfoster 3 months ago (Flag)

hmm - good point that is weird I'm not sure why https://www.inaturalist.org/observations/38659843#activity_identification_bf54de81-0033-4642-ad00-999f648724a7 was coarsened based on that location. There don't seem to be a ton of other examples though (or at least examples that haven't been refined manually) https://www.inaturalist.org/observations?locale=en&lrank=genus&preferred_place_id=1&subview=map&taxon_id=7902 I wonder if there was something weird with https://www.inaturalist.org/observations/38659843 e.g. like its location was changed and somehow didn't get properly indexed or something. Odd

Posted by loarie 3 months ago (Flag)

From what I can see all records from within the below encompassing place were affected. I think it must have inadvertently been included in the atlases of both species when the split was committed. Looks like the records have mostly been tidied up manually, now.
Oceania (Continent)
Australia (Country)
South Australia, AU (State)
Adelaide Hills-Bal, SA, AU (County)

Posted by rfoster 3 months ago (Flag)

Add a Comment

Sign In or Sign Up to add comments