IDs on species complex rank should automatically reach research grade

Feature Requests
1

Platform(s), all

Description of need:
Describe the iNaturalist community need that your requested feature addresses. Include screenshots, URLs, and other details to help us all understand the issue.

The species complex is established in cases where similar and related species are barely distinguishable. The aim is the use the complex level for research because there only poor data on the (micro) species. Like subspecies in most cases we can’t expect of the species with a more or less random photo. I sometimes use the option „cannot be improved“. However I do not think this should be done manually. With that complexes should be treated as if they where species with subtaxa, similar to species with subspecies, variants or forms.

Feature request details:
Examples I was looking into:
Portulaca oleracea has 46.031 observations. However it turned out to be considerd as a complex. Without a very diligent examination of the seed surface, one cannot tell apart (aside from genetics).
In practical terms this means it would be accurate to move the 46.031 observations of the species to the complex, telling GBIF that the complex moved from very common to almost extinct.

(1) (PDF) Microspecies of the Portulaca oleracea aggregate found on major Mediterranean islands (Sicily, Cyprus, Crete, Rhodes)

The complex Arocatus roeselii ist spreading under Platanus bark in europa. There is some scientifc discussion about what is going on there. However we don’t share our knowledge with science, because it is just unclear, weather there two species or may be even something in between two formly separated species.

“In the last years, the limits of A. longiceps and A. roeselii became uncertain, because the specimens found in Western Europe on Platanus showing the characters of A. roeselii together with typical longiceps specimens and some transitional exemplars (Carayon, 1989; Barclay, 2007; Hoffmann, 2008). Hoffmann (2012) tried to fi nd at least genetic difference between the both species but it was unsuccessful. Therefore the validity of A. longiceps is questionable. However, when check the genitalia of them, we fi nd the pygophore opening is parallel in anterior part in A. longiceps, whereas anteriorly widened in A. roeselii (Fig. 2A, D). In addition, parameres are also different, e.g., base of blade nearly straight while in A. roeselii it is strongly convex (Fig. 2B–C, E–F). The decision needs further investigations.”
https://lkcnhm.nus.edu.sg/wp-content/uploads/sites/10/app/uploads/2017/06/61rbz687-704.pdf
https://ukrbin.com/downloads.php?catid=0&getfile=10

==

There are some 30000 observations of the complex Geranium molle.
21 (!) of those are identified not to be Geranium molle s. str. but one of two others in the complex

Some 14000 didn’t get research grade mainly because iders know that they would need specific traits to identify that almost never are visible on photos unless the observer explicitly searched to demonstrate them. And from those that are identified to the species a rather great part is not really qualified. However almost nobody likes to destroy the GBIF transfer with a hard disagreement. On the other hand: no scientist would expect to get Geranium molle from iNat or any other platform because it’s clear that a great part of observations would be good enough only for the complex.
To mention it: some 700 observations currently are at the complex level.

==

Oulema melanopus: Without dissection you can’t tell the species of this complex apart. More than 4300 Observation are identified to the complex and don’t get and cannot get research grade, without marking as cannot be improved whicht would not add much information.

==

Spiranthes cernua
Almost 16000 observations in total, 13 species, almost 1200 stick in the complex but would be of course interesting eg for conservation

==

Hyla versicolor
2 species, over 28000 observations, more than 12000 observations als species complex

==

Entomophthora muscae (a parasitic fungus)
About 4300 observations in the complex. While the host might help to identify might help, this is not perfectly clear. Expert tend to be reluctand.

==

I would to leave it with those examples.

2

Since there are some exceptions to the above, particularly that ‘needs ID’ is more likely to achieve further IDs and this is sometimes possible, and things staying at ‘needs ID’ when they are not is a problem in many cases where species complexes haven’t been formally created, I’d rather see an improvement to the “Based on the evidence, can the Community Taxon still be confirmed or improved?” feature that makes it more likely and convenient to be used by identifiers.

8 Likes
3

I would be against this. For two Complexes I encounter quite a bit, they are only “unidentifiable” in small portions of the overall range. However, in both cases because the Complex name is the same as the more well known species, and in one case, most people are unaware of the ranges, we end up with a lot of observations ID’d to Complex that are certainly easier identifiable.

In both cases, some of us are working on boiling them down further. So this change seems like it would be very Complex-by-Complex which I’m not sure the site is really set up for.

10 Likes
4

It is probably worth mentioning that I believe observations at the rank of ‘species complex’, even if they are RG, are not exposed by GBIF in that form. The GBIF taxonomy deals only with formal ranks. The iNat observations are exported/imported with identifications at the next higher formal rank. So, they do go to GBIF, but with loss of precision.

9 Likes
5

Yes, usually GBIF equates iNat’s “complex” with the Genus rank.

Because there is a range of different yet sensible meanings and uses for “complex” on iNaturalist, automating/systematizing action around that half-baked rank might cause as many problems as it solves. So… I agree that “IDs on species complex rank” are best dealt with on a case-by-case basis.

3 Likes
6

This is a very common case, and people often select the complex instead of the species because it comes up first in the list. For example, the ubiquitous (in North America) “Bold Jumping Spider” Phidippus audax - is the nominate species of the audax “complex.” So when you type “Phidippus audax” the first thing that comes up is “Complex Phidippus audax”. Many people click this without realizing they have selected a species complex and not a species. It would not be good to automatically make all of these RG. As noted, while some closely-related species are truly cryptic, there are many complexes where species can be determined by range, somatic characters, phenology, etc.

2 Likes
7

This has been proposed before and staff has stated they don’t plan to do this: Make observations with a species complex level community ID automatically become Research Grade

4 Likes
8

As I made the previous request I support this, but there are some issues that would need to be addressed first that were brought up in that discussion (and have already been brought up here). That includes that fact that species groups don’t exist as an option yet. Generally most species complexes sensu stricto will not be identifiable. But currently we have an expanded concept that includes many groups that should really be classified as species groups, and species in those groups are often identifiable.

2 Likes
9

What I’d find really useful, to speed up identification of moths, would be a button which, with a single click, both added an id at the next higher level than that given (whether a “complex” or a genus) and also ticked “cannot be improved”. We could then very quickly sort out a lot of observations which have been identified to a species where this is not possible from a photograph alone. This could always be over-ruled later by the normal means if a species id becomes apparent.

1 Like
10

I can see the need, but it’s quite simple to mark DQA community taxon status ‘as good as it can be’, which helps avoid situations where an organism can be further IDed being left at the complex

3 Likes
11

Well, that is indeed the strongest argument against my proposal. I did not realize that GBIF doesn’t support complexes. Push those complexes to the Genus level doesn’t make too much sense. I guess I will continue to ignore some of the complexes and ID to the main species of the complex.
To keep those ID’s accurate it might be helpful to have a data quality option “sensu strictu” or “sensu latu” instead.

12

Strong oppose, as someone who routinely deals with difficult to identify species (ants) in groups and complexes, there are plenty of cases where these are identifiable to species on iNat, even in sensu stricto complexes. Many groups or complexes contain both a very common species and other rare or newly introduced species, removing these observations from the “Needs ID” pool would have a catastrophic effect on iNat’s ability to detect rare species, range extensions, and new invasives

I do think the “good as can be” button is underutilized in cases where something can only be IDed with a macro lens, and the observer doesn’t have one, but the solution cannot be to move the obs that do use a macro lens out of the needs ID pool

6 Likes
13

Surely making inaccurately precise IDs is worse than being unhelpfully imprecise?

That’s exactly what used to happen in the iNat → GBIF transfer and it was corrected: Issue With Species Complexes/Groups in Relation to GBIF

5 Likes
14

You can simply ID to complex and mark the community taxon status in the DQA ‘good as it can be’ which will allow observations to get to research grade

3 Likes
15

If doing so means the observations would be summarized under the genus, the button doesn’t help much.
I am honestly not aware of an example, where species within a complex can be distinguished easily.
Here are my examples, where I tend to ignore the complex:
Draba verna ( Bomble: Draba subgen. Erophila in Deutschland. Auf dem Weg zu einer natürlicheren Taxonomie)

87 Observations as complex:
Beobachtungen · iNaturalist

45.937 Observations for the species where in practical terms none is qualified to be distinguished as one of the micro species, as they are used in publications like:
(1) (PDF) The current knowledge about the distribution of Portulaca species (Portulacaceae) within the Balkan Peninsula
Bomble: Draba subgen. Erophila in Deutschland. Auf dem Weg zu einer natürlicheren Taxonomie

1 Like
16

The observation are not summarized under a genus, they are summarized as the complex unless I am misunderstanding what you’re saying. And yeah the technical definition of a complex is simply closely related species that are hard to differentiate, more or less (exact definition often depends on situation and what branch of life etc.).

1 Like
17

My understandig: complexes are not transfered to GBIF as complexes. So it is not possbile to select GBIF dots for a complex nor to find complexes on GBIF itself.

image
image235×261 22.5 KB

image
image728×591 51.4 KB

I guess the simple reason is: complexes are not part of the official taxonomy but just a construct for practical reasons in cases where data to correct species are barly available.

In other words: while there is some scientific literature dealing and defining complexes, they are not and can not be valid taxons.

1 Like
18

I found one: the complex Centaurea jacea includes two species and a hybrid. While C. jacea is widespread an can be distiungished easily, the other two have a very restricted area. In by far most cases it is save to ID the species sensu strictu.
Komplex Centaurea jacea · iNaturalist

In by far most cases no IDer would even consider the complex.
That said it does not make much of a difference related to my initial idea.

19

Please do not ID observations beyond your ability to make a definitive ID. Not every observation needs to or should reach RG at the species level. As I often tell the biologists I train, incorrect data is worse than no data.

Not to be pedantic, but this isn’t the definition of a subspecies. Many subspecies are relatively easy to differentiate visually. Furthermore, species complexes often represent different species (not subspecies) that are closely related and difficult to differentiate.
Update: Text corrected in quoted post. Poster meant “complexes” not “subspecies”.

4 Likes
20

lol that should have read complexes and then it would have made a lot more sense. I was probably just discussing subspecies recently and had them on my mind (big issue in the herpetology world).

4 Likes