Yes, I guess many of those from iders who check unknowns too, maybe there should be more materials posted on forum about iding plants? There’re so many tutorials on animals, but none about big plant groups, in 99% of cases you don’t need to know the plant species to id it lower group, especially when plants include mosses and part of algae, so this high rank shouldn’t be used that frequently.
5 Likes
On the one hand, you’re absolutely right, it shouldn’t be too hard to learn a few lower ranks. On the other hand, when I ID plants I most often use the iconic taxa button and a geographic filter, so it really doesn’t matter to me if something is Plantae or family level, I’m going to see it all. So it kind of depends on whose attention you’re trying to get and whether “dicots” will really gain you any faster attention than “plants.”
5 Likes
I know many iders don’t know mosses, so they filter up to Tracheophyta.
I can testify to this. As a novice IDer I started going through Unknowns and I admit I marked quite a few as Plantae. I used “Flowering Plants” only for plants that were, well, flowering (since that seemed pretty safe) and never “Vascular” because who knows what that means. I still don’t know exactly what vascular means, but it seems to be most “normal” plants, so maybe something to do with having a stem. Eventually I learned to use Vascular and Flowering properly, and now I’m working on Dicot vs Monocot. But it would have been a lot easier to figure out if there were some sort of ID handbook (just info on the lowest levels you can easily ID things to would be great!) I’ve found some of this information on the forum, but it would be nice to have it consolidated as an iNat page. I think it could cut down on the not-very-useful kingdom-level IDs.
7 Likes
Vascular plants are pretty much all of them except for mosses, liverworts and hornworts. It includes all flowering plants, gymnosperms (e.g. conifers), and the ferns and fern allies. If it has an upright stem, particularly a woody one, and leaves with veins, it’s a vascular plant.
The non-vascular ones are commonly called bryophytes, but on iNat the taxon Bryophyta is mosses only. Based on having gone through the “Bryophyta” in my area, it appears this isn’t clear to everyone and there are a lot of liverworts marked as such as well. If anyone who can distinguish between a true moss and a liverwort or clubmoss is interested in helping with sorting these out, I suggest having a look at Bryophyta in your area and correcting things that are obviously not mosses. I found a good number of them.
I agree it would be really nice to have some quick references/guidelines for sorting out plants at least to phylum level or classes.
7 Likes
I’m embarrassed to say this but I actually didn’t realize liverworts aren’t bryophytes and I’m a plant ecologist 
9 Likes
Well, they are “bryophytes” but not “Bryophyta” as iNat uses it (for mosses only) - that’s where the wrong IDs come in. I’m sure it confuses a lot of people!
5 Likes
I do a lot of coarse id’s for the plant side from Unknown, and it comes down to workflow convenience at high volume. Plantae comes from either that kingdom-based app, or from the Identify pane if “P” is a good place for me to repetitively tap. Angiospermae is a better high volume choice because A fits well into my keystrokes and covers most things (since we’re seeing mostly monocots and dicots, and no big need to make a choice right then). D is likewise handy/easy tap for long runs of dicots, M slightly less handy so I often default back to Angiospermae for a while if I get tired of doing M in the mix.
The curveball in the high volume plant id workflow is when I try to rely on my touchscreen after the keystroke. After a certain amount of ids, the website starts to glitch and fill in random not-selected id’s from the pulldown like Duck or Diptera or Anisoptera instead of Dicot, until it somehow reverts to normal after a while. (Thread about that is elsewhere here iirc).
For “normal people” doing plant id though, just go with your personal comfort level, as always. ;)
7 Likes
In my dry climate there’s very little moss anyway; almost no one observes it. I spend much more time ignoring obs of seaweed but even those are not all that many.
1 Like
I wonder if theres any way to get statistics on how fast things get cleared from various levels, or how many IDers work there? It seems like most stuff that gets a starter ID of tracheophyta is something like a small patch of bark or sliver of wood where someone looked at it and couldn’t decide whether or not it was an angiosperm. On the other hand especially in dry areas almost everything can be progressed beyond plantae starter IDs relatively easily, so it might feel more productive to ID from there if you are going to filter by a specific higher ID level.
4 Likes
I don’t know if it’s possible for the average person to access info like that. I know the site stats page displays graphs about time to ID or to CID, but it’s not filtered by taxon.
1 Like
I’ve been going through a month’s worth of Unknowns for New England in the US (yes, I’m a masochist) and I’ve been marking as Life observations with several photos, each clearly of different organisms. If no one has commented before that these photos should be split into different observations, I make that comment.
But in cases where an identifier asked the observer a couple of months ago to split up the photos and the observer hasn’t responded, I’m thinking of marking the observation as Life - and marking it as cannot be improved. What do you all think of that? It would, I think, move that observation out of the Needs ID pile.
11 Likes
It’s a right actio and really it needs to be done right away, with addition to the comment about split you’d like to add to observer to ask them to comment when they will do the split. Id should be that of common ancestor of all oranisms on photos, so choose it and mark observation, there’s a pattern of other people not interacting with observation after such comment, so it can stay in Needs id for years.
4 Likes
That’s a good point about choosing the common ancestor of all of the photos. I’m inclined to give the observer at least a couple of weeks to notice that someone made a comment on their iNat observation, rather than causing it to be moved out of the Needs ID pile, but yeah, there are way too many multiple-organism observations that linger for years.
(Am I the only one that feels the need to finish dealing with my local subset of Needs ID observations? I know that’s impossible, but I’d like to feel as though we’re making a dent in the ever-growing pile.)
3 Likes
I’ve done that with a fair amount of photos of groups of turtles basking. Sometimes they’re all one species (usually red ear sliders, sometimes a cooter), and a lot of times people comment on which turtle they’re referring to…but there’s also a lot of photos that are a mix of various gregarious baskers…
1 Like
Be careful of a pitfall there. I have seen Araceae with broad, net-veined leaves get stuck at “Flowering Plants.” The first reviewer assumed they were Dicots, because their leaves were not grass-like, then the second person (sometimes me) correctly IDed the family Araceae, which is a Monocot family.
In California, I see another frequent mistake, namely, that Ranunculus californicus has unusually many petals for a buttercup, so some people think it is Asteraceae.
3 Likes
Don’t worry, I always follow up on disagreeing IDs and generally withdraw mine if one is added.
6 Likes
100% true. It isn’t right just to take and never give back.
I just today implemented what i hope may alleviate this problem. I figure, a certain percentage of users are using the CV for taxa they do not know. I noticed that the default picture for Achyrachaena mollis is its distinctive seed head; but the default picture for Uropappus lindleyi was its flower head. So, if the observer is uploading an observation of the seed head, and sees the CV suggestions, they would notice that Achyrachaena mollis looks kinda similar, but would not see any similarity in Uropappus lindleyi.
Therefore, I went to the taxon page for Uropappus lindleyi and set the default picture as one of the seed head. That way, when someone sees the CV suggestions, they will be able to compare the two seed heads, and hopefully see that one looks more like their observation than the other.
I once heard an educator refer to this sort of thing as “structuring for success.”
18 Likes
I agree! I edit or add taxon photos often, making sure that the first photo is the most typical of what you’d see in the field, and that later photos include diagnostic characteristics, a scale object, and multiple life stages if possible. People at all skill levels use those images constantly, and new users are often confused if the taxon photo and their specimen don’t match.
9 Likes
For anyone interested, this is from iSpot.
When one posted an ID one could choose (Confidence):
- It might be this
- It is likely to be this
- I am as sure as can be.
(As a rule beginners are “more sure” than more regular users - it takes experience to discover how little you know).
And iSpot had a reputation system. Taxonomists and Museum curators and Monographers got a reputation of 1000, and experienced people got a reputation of 500. Normal users got a reputation of 1 and if they made IDs that others agreed to their reptutation increased to a maximum of 500. As a rule it takes over 1000 correct IDs to earn a reputation of 500.
What iSpot did is summed up the reputation of identifiers and agreers and came up with a score. The top scoring ID got the “Likely ID = True” and the remainder of IDs were “Likely_ID= False”
What was cool about the reputation system was the “Expert Equivalents” : that is the sum score of the reputations of everyone who identified and agreed.
Basically 1 EE = one museum taxonomist level. So an EE of 5 would be excellent, and an EE=<1 would be one where no expert had reviewed it, but 0.5 would be a resonable ID (compared to say 0.001 for a beginner ID with no verification).
So in this case the ID was 0.08 expert equivalents - not very good, so not surprizing that the ID was False. As for the Confidence - as I said, beginners would go to the field guides, make an ID or a near perfect match, be extremely confident (and probably quite proud), and only after a few dozen IDs would discover from some other users that the field guide illustrated one species out of 15 in the genus, and 15 other similar genera existed that were not even mentioned and the taxonomy was last reviewed in 1905, by a European taxonomist who never even visited in Africa, and that at least 20 more species were waiting to be described, etc. etc.
I still think the system is far superior to that of 1 ID by anyone and 1 Agreement = Research Grade.
Although if you download data, I recommend that you use the fields.
- num_identification_agreements
- num_identification_disagreements
as a proxy for “confidence” in the Identifications. Except that just like the iSpot reputation system, some groups have lots of IDs and some hardly any, so that one needs to be careful in comparing different groups (like Storks vs Cisticolas).
15 Likes