Issue with Class Reptilia Data Download

[gbif-portal]

Issue with Class Reptilia Data Download

Upon accessing the downloaded data, I noticed that the Class Reptile records appear to be missing. Instead, it seems that the columns that originally belonged to the Class Reptile have been reassigned to a new class. This discrepancy has raised concerns regarding the accuracy and completeness of the dataset. I kindly request your assistance in rectifying this situation. It would greatly benefit the research community if the Class Reptilia occurrence data could be restored to its correct state, with the corresponding columns and classifications properly assigned. Thank you for your attention to this matter. I look forward to hearing from you soon. Best regards,

---

User: See in registry - Send email System: Chrome 115.0.0 / Windows 10.0.0 Referer: https://www.gbif.org/contact-us Window size: width 2560 - height 1283 API log&_a=(columns:!(_source),filters:!(),index:'3390a910-fcda-11ea-a9ab-4375f2a9d11c',interval:auto,query:(language:kuery,query:''),sort:!())) Site log&_a=(columns:!(_source),filters:!(),index:'5c73f360-fce3-11ea-a9ab-4375f2a9d11c',interval:auto,query:(language:kuery,query:''),sort:!())) System health at time of feedback: OPERATIONAL

[khanhluongpy]

Issue with Class Reptilia Data Download - crop screen image

[ktotsum]

contacted the user from helpdesk on 2023-09-08.

The status of Class Reptilia is proparte synonym and therefore occurrence data of Reptilia are indexed to four classes (Sphenodontia, Squamata, Testudines, Crocodylia).

You can search these classes as follows: https://www.gbif.org/occurrence/search?taxon_key=11569602&taxon_key=11592253&taxon_key=11418114&taxon_key=11493978

[khanhluongpy]

I'm sorry, I know this. But Order field is missing data.

[ktotsum]

@khanhluongpy, Apologies for missing the point here. Thanks @MattBlissett for the follow-up!

[ManonGros]

There is no order any more in the catalogue of Life for these classes:

the GBIF backbone doesn't support intermediary ranks (yet) so the other ranks aren't shown.

[mdoering]

Yes, this is unfortunate in COL I think. There is an ongoing taxonomic discussion within COL about how to deal best with missing major Linnean ranks.

https://www.catalogueoflife.org/?taxonKey=45C https://github.com/CatalogueOfLife/testing/issues/186 https://github.com/CatalogueOfLife/data/labels/management%20hierarchy

[khanhluongpy]

It's been a while since I reported this issue but there hasn't been a new fix update yet. Image:

[mdoering]

We completely rely on the Catalogue of Life to organize the reptiles. Since we last built our backbone COL has moved the 4 reptile groups from class to orders under a Tetrapoda megaclass.

I don't think you will see the paraphyletic Reptilia being resurrected as a single class. With the next backbone update in GBIF you would then see orders without a class. We are working at GBIF to include more ranks so then at least you can then see the Tetrapods megaclass, but that will still take a while as it is a big change for us.

[NicBailly]

In reference to Markus' email about an user missing Reptilia.

• I think for the search facility, we have solved the case with synonymies pro parte, as already explained in a message above.
• If I understand between the line, the user misses the name in the hierarchy.

There we do not have many options:

• one that we DO NOT have is keeping the name in its old circumscription as said above. However, our current solution is only a halfway solution since the 4 orders are gathered in a placeholder class, which phylogenetically should not be. We just removed the name. One issue there is how much the inclusion of fossils will affect this part of the tree.

• uplift the 4 orders at class level, which creates the problem with the GSD because the anchor points would be at different level. We could think for the moment to add the four classes, Crocodylia-class, Rhynchocephalia-class, Squamata-class, Testudine-Class, that would contain each one order only, the orders that match the GSD so the GSD does not need to change anything, and people searching Crocodylia only would get what they expect not bothering about the rank: whatever they are looking for or they will get the same (again fossils may affect that solution, the question being restricting these names to extant species or not). But it does not help to see Reptilia in the tree.

• reintroduce the name with the modified cladistic/phylogenetic circumscription as a superclass including Aves, which forces to create two placeholders at superclass level for Amphibia and Mammalia.

• we could think to trick the display by adding some names. But this would create a lot of patchy programming work I guess.

• we could think also to use the names Aves-Reptilia, Crocodylia-Reptilia, etc., but it will affect the search capacity

My suggestion is that we link concerned taxa to a file that:

• Would explain the difficulty to have a full classification at that place of the tree because it would multiply the ranks;
• And that will contain a more precise hierarchy of all Vertebrata with more details (and remember there will be adjustments because of fossils), with useful links to recreate the sets organised hierarchically as people want to use them, in GBIF and COL. This solution could be generalized to other part of the trees.

And the conclusion that I expressed many times fro more that a decade is that we have to be creative for new ways to display classification: I will add now, as long as it requires only a few explanations, it is intuitive, and data operation are kept manageable for people with a reasonable knowledge of spreadsheets (e.g., a hierarchy with reasonable number of ranks). We could look to the Phylocode and implement some of the solutions they propose: in general, I am not favourable of a complete use of the Phylocode, it is not practical by non-taxonomist people, but one cannot deny that some idiosyncrasies of the Linnaean hierarchy could be lessen by adopting some ideas.

[mdoering]

My main critic is the missing class for the 4 reptile orders. I think COL should strive for a complete classification using the major Linnean ranks. Additional ranks are very fine, but the main ones should just never be empty