Flag for Taxon: Common Flesh Fly (Sarcophaga carnaria)
| Flagger | Content Author | Content | Reason | Flag Created | Resolved by | Resolution |
|---|---|---|---|---|---|---|
| treegrow | Common Flesh Fly (Sarcophaga carnaria) |
This species does not occur in the Americas, but there are lots of New World records due to misleading iNat ID suggestions. |
Apr. 13, 2020 04:26:20 +0000 | zdanko |
under control |
Comments
We have several sarcophagid genera that look similar to Sarcophaga, so it's not safe to ID sarcophagids to genus unless you know how to tell them apart.
I've just booted out about 28 observations from the North American region.
To help prevent this from perpetually occurring, I've updated the default name for North America to its more appropriate name. It seems part of the issue is that only S. carnaria pops up when searching for the phrase "flesh fly", so I've also made sure the family name will now pop up in the ID box. There's also another flesh fly with a common name that will now pop up. It seems most (~99.9%) flesh flies don't have species-specific common names, so there was a sort of false positive.
Thanks, I occasionally go through the obs for this species to clean out the American records, but the pace of replenishment seems to have gotten faster lately. Changing the vernacular may help, but I suspect the bulk of the misidentifications is due to iNat autosuggestions. Maybe it's coming up more frequently in the recently retrained model.
Looking at what range data I can find, it looks like several other regions with out-of-range records. The issue that only males are even identifiable to species, and even then requiring microscopic examination of genitalia, makes it seem that even within range, species-level IDs aren't reasonable.
Yes, it's one of these cases, where most if not all species IDs are unreasonable, but we don't have anybody knowledgeable enough to lead a well reasoned clean-up. There are many fly species where this is true, but it's really hard to convince people to stick to genus or family when the autosuggestions provide them with a species name or they find a picture on BugGuide that "looks just like it."
fixed the Australian + Se Asian ones
I'm attempting to reconstruct a usable atlas for this species. I've cut the tentative Russian range at the Altai Mts. as of Pape (1987). Occurrence data as of PESI and Fauna Eueopea are a bit more restricted remove Russia entirely. All seem to agree that India, Pakistan, Spain, and Portugal all fall under out-of-range territory. 20 observations are currently outside of the atlas (22 if Russia is excluded entirely).
• http://www.eu-nomen.eu/portal/taxon.php?GUID=urn:lsid:faunaeur.org:taxname:142340
• https://fauna-eu.org/cdm_dataportal/taxon/362c3913-6248-4ec9-9b70-2824679b3090
Many thanks to all of you - I appreciate the efforts at changing the common name. I posted this as a forum topic, but generally got shouted down... :)
https://forum.inaturalist.org/t/changing-common-names-common-flesh-fly/11255/26
Jesus, in three days 15+ more just got ID'ed in the US, Australia, etc. Have removed those again
CNC has been a killer for these, I've fixed 75+ in the last couple of days
As far as I know they are not identifiable to species in Europe either, so we should be good to move European observations back to family as well? That would take the species out of the CV suggestions in the next update.
I'm up for it if you want to take the lead. We definitely need a posse to knock them all back. There are quite a few that have reached research grade. I'm happy to be tagged whenever additional IDs are needed.
I tried to help out a bit with this, bumping many of these species back up to the family level. Despite the bug (computer bug) being reported about the change, it seemed like a few of these observations fell through the cracks. I did a copy/paste message on most of the observations with this flag dialogue.
Thanks for pitching in. It looks like the US/Canada is clean now. It will be interesting to see how long it takes to fill up again.
The rest of the Americas, Oceania, and Africa are re-cleared as well now.
The continent of Asia has been re-cleared of all but 1 record (needing an additional disagreement).
At some point, Europe probably has to be tackled (I'd suppose beginning with RG records since that influences CV). From the literature, a species ID is definitely impossible apart from dissection of male genitalia (and I'm only aware of 1 record in the whole carnaria-group that has this requirement).
We should now be down to a single RG observation (the only one that involved dissection of male genitalia). This will hopefully have a significant impact on the accuracy of Computer Vision.
This presently leaves 469 "needs ID" records (of which it looks like all but ~4 from a single user are erroneously diagnosed).
Hi guys, thanks for your efforts so far to correct the IDs. I noticed that many disagreements (at least in Europe) did not end up in correcting the ID. Probably a bug, so I made a report in the forum: https://forum.inaturalist.org/t/disgreement-with-wrong-taxon-level/19148
Initially thought this was a single weird case, but found a lot more in the meantime.
@chrisrap @edanko @monicaoyre @rui_andrade @entomokot @waldgeist
The disagreements all intially worked as normal and corrected the ID; however, when the S. carnaria complex taxon was created, the disagreement was lost for some reason (I had to re-correct all the Australian ones). I filed a bug report in November here: https://forum.inaturalist.org/t/previous-disagreeing-id-erroneously-overturned/18183
That's annoying that so many observations have to be re-IDed, but I am willing to help cleaning this up, but I would like to gain some more knowledge with this taxon, so I'd like to ask this:
As there are now many observations of the S. carnaria-complex, maybe someone can confirm that the following combination of characters allow a classification into that group (consisting of S. carnaria, variegata, subvicina):
Last abdominal tergites black + 4 DC + marginal bristles on 3rd tergite?
And a second request for clarification: do all sarcophagids with three black thorax stripes belong to Sarcophagini?
I've never heard of being able to ID Sarcophaga-like observations to genus based on field characters, let alone species, that's something I'd need to learn more about. There are some European fly IDers who claim they can ID to genus Sarcophaga, but as @treegrow said more work is needed before anyone can say the same in the Nearctic. Maybe @waldgeist would know more (I don't think he's one of the ones who IDs to genus)
For your second question, there are dozens of genera in Miltogramminae and Paramacronychiinae with three black thorax stripes.... three black thorax stripes and a pale basicosta is apparently the ancestral state of Sarcophagidae .... there are even some Calliphoridae that look almost identical.
https://www.inaturalist.org/observations/37500032
No, I cannot id "flesh flies", I try it only on Miltogramminae
Eurychaeta ("false flesh fly") is also a Calliphoridae genus with sarcophagid-look-like flies
@carnifex you can quickly separate most of the Sarcophagid look-alikes by checking for the pale basicosta -- the look-alikes that also have the pale basicosta tend to have a wildly different abdomen
Sarcophagidae key (british isles) https://osf.io/vf5r6/
"Key to adultflesh flies (Diptera: Sarcophagidae) of the British Isles
DANIEL WHITMORE, STEEN DUPONT, STEVEN FALK"
Thanks, I found this statement (three stripes = Sarcophagini) as a comment under one observation and was sceptical this might be useful.
I was mainly working with the British key linked by @waldgeist and with DREES (2014) 'Die Fleischfliegen des Hagener Raumes', where I also found the criteria for assignment to the 'carnaria-group'.
I guess one has to thoroughly check IDs from this taxon as well. Anybody knows if there are more than these three species to be included in the carnaria-complex?
Meanwhile, only 6 European countries left with uncorrected S. carnaria observations :-)
jeepers so many of these popped up during the CNC. Removed 70+ today
As expected :-)
I didn't have time to check, but I assume the same is true for Drosophila (melanogaster) and Condylostylus patibulatus.
Please let's combine forces!
I corrected a few D. melanogaster and C. patibulatus from my western NA region yesterday...
Here's one that's likely out of range: https://www.inaturalist.org/observations/79123892
India's popping up green on the map, but shouldn't it be absent from there?
@cobalt317
In 'The Flesh Flies of Central Europe' 1997 it mentions only Russia as an Asian country, so it should not occur in India
Hey guys,
I would've normally put a flag on family level, but as here are already so many knowledgeable people involved, I rather continue the discussion: I feel the iNat taxonomy of the family is quite a mess and needs to be cleaned up. So I would suggest to follow a certain authority. I asked Daniel Whitmore (British Sarcophagidae Recording Scheme), and they refer to http://diptera.dk/sarco/index.php (by Th. Pape). It dates from 2014 but is on expert level and very much complete. Because otherwise, it might become chaotic especially at (sub)genus level of Sarcophaga. I also propose to simplify it by not using the Tribus level, which is also not commonly used in this group and rather vague. So, @jonathan142 and others, what are your thoughts on this?
That sounds like a good plan, but it looks like a lot of work. Let us know if you need help with anything.
Don't know when I find time to tackle this, first wanted to ensure to not step on somebody's toes
According to the new CV model, the status on the species page of S. carnaria is indicated as 'pending' - would that mean less CV suggestions in the future? Here's hoping it will turn out to be the case...
...apparently it did, just the Seek App won't stop suggesting that species
I've removed the erroneous atlas listing for India in the atlas. But there's no user listed when it was added (?)
It does look like CV is starting to suggest the species again, but at least not as prevalently.
I don't feel too strongly either way on tribes, though there are authors who use them. Most users who add the species ID don't seem to be looking at the taxonomy at all anyway, so it really isn't helping to distance users from an implausible or impossible ID. Plus, most disagreements are at the family level (or occasionally subfamily level).
"It does look like CV is starting to suggest the species again" these are all from seek unfortunately, none from iNat itself
I'm going to resolve this flag, as the situation is rather under control. There were very few observations when I looked, and I went through the 10 without dissection and put them back to family. If this species starts becoming a problem again feel free to unresolve.
Would American observations also be Sarcophaga (for the ones that actually are Sarcophagidae of course)?