Is it possible to count how many plants are IDed directly from Unknown to Family or lower?
Versus how many limp to Family via Plantae, Trachaeophyta, Angiospermae, monocot or dicot?
The major part of iNat obs is plants. When IDing if I can’t get to family I leave them at Unknown. I presume (from my ignorance) that bryophytes and conifers have a reasonable chance. I know Fungi and Lecanaromycetes get IDs quickly.
How many identifiers filter for those very broad plant categories?
Does it make a tangible difference to IDing plants if I would add those ‘almost unknown’ IDs above family??
It definitely makes a difference. I frequently ID plants from unknown to Angiospermae, monocot or dicot and regularly see them IDed to genus or species the following day or soon thereafter.
Just to add… I am just beginning to id other people’s photos. I look at State of Life/ Unknown because I’m new, but I can sort things into big bins. I try to work from things posted longer ago. It’s very rewarding to see updates that within days something’s been ided that languished in State of Life, sometimes for +year. Any push to get it seen by better people is a good thing. ( even when I’ve been embarrassingly wrong!)
It is not possible. Blue and I have this problem every identathon–we’d really love to be able to exactly count how many unknowns were identified, but you can’t simply because once they are identified, they’re not unknowns. API calls ask for the current status of observations, not past status. And also there’s no such filter for “number of IDs on an observation=1” or “first identifier is not the observer.”
That being said, absolutely you should ID unknown plants to kingdom. I filter for kingdom Plantae every day and I have multiple helpers who put the plants in there from the unknown pile. At least one of those people just puts “Plantae”, and yet the two of us together have identified 2,019 observations, of which 1,217 are currently at species.
I do! I frequently start my ID sessions with picking out the plants I know from the Unknowns, then switch to plants at Kingdom and Phylum level, and then to everything from Class to Family. Sometimes I’ll look through the species level IDs for my immediate vicinity to help knock them to RG or correct misidentifications, but mostly I look through the broader categories. It helps split up the huge pool of plants needing IDs into reasonable chunks over larger areas. Below family level, the pool gets so big that I then have to cut it down by choosing smaller areas, going county by county or even just focusing on one State Park or National Park.
Some kind help please for the Rest of Africa Plants. If I have built this URL right? - plants down to family, needing skilled identifiers, 35K obs waiting hopefully.
If the ID is ‘Unknown’, I always tell new users to stick in a rough ID, as Unknowns tend to get lost in the shuffle.
I have seen a number of observations quickly go from that to species in a short amount of time. If that’s not what you meant, apologies from a moth person!
I am convinced now. 8 out of 10 obs yesterday evening were … hmm … Dicot. Was afraid of lots of notifications today, but not. Will be interesting to see my responses.
I am also willing to step out of my comfort zone and push a hook line sinker response of - no it’s not that.
My philosophy on this is a combination of “Every Little Helps” with “Do What Makes You Happy”. If you have the patience to push some observations from Unknown to Monocot, please know that you’re adding information that may well reduce the work for other identifiers. And if today you don’t feel like adding those coarse IDs, don’t feel any obligation.
I mostly focus my IDs on combinations of taxa and countries where I have some knowledge. Once I’ve trawled through all the observations within a particular genus, I’ll often broaden out my search in progressive steps: all Agavoideae with IDs at subfamily level; all Asparagaceae with IDs at family level, etc. Eventually, this gets up to all Angiospermae, Tracheophyta, Plantae, etc, that are not identified more precisely. Once in a while, I’ll dip into Unknowns as well. Despite this, I do notice that I pay a lot more attention to searches focused on the lower and middle ranks and I run these more often. So any identifier willing to kick a few observations a rung or two down the hierarchy is helping me out when I go to perform more detailed ID work.
But that really shouldn’t mean that anyone should feel obliged to add these high-level IDs. Certainly, I regularly skip over pages and pages of plants that I could ID slightly better because I’m choosing to search for a particular genus or species and don’t feel like doing the repetitious work of moving hundreds of observations into Dicots or Asteraceae.
I think any level of ID can be useful for two reasons:
It will increase the chance of some expert or enthusiast to pick it up and have a go at identifying it further.
Many recorders on INat are enthusiastic amateurs who are there to learn more about what they see and record. Any level of ID beyond unknown or plant can help them a bit in their personal learning curve.
.
I would suggest you change per_page=10 to a larger number. As an identifier, I don’t like having to switch pages often, because it disrupts my rhythm. I often sort the high levels with per_page=100.
I did a small number of them tonight. I can’t do many fine IDs, but I did get a few to family or class.
I look through Dicots often. Not just because people identify unknown plants to that, but also because two conflicting IDs often puts the plants in Dicots.
I don’t generally do monocots because I’m not as familiar with them, but I sure hope someone is. Conflicting IDs often put things in these broad categories, and they can sometimes be easily resolved. A conflict between a conifer and a dicot will put something in “Vascular Plants” for instance.
There are two families I look through for my area, and one subfamily, for a few particular species that I’m comfortable identifying to species level.
On the flip side, when identifying coarse level observations or unknowns, I often use the subfamily “Amygdaloideae” for things that look like cherry/apple/etc. I often think I should stop doing that because it seems like none of them make it past subfamily, so I’m not sure if anyone is looking at that group of plants at all…
No thank you, Becky. I have set it at 10 because it fits my screen. I can scan 10 - sort into Agree. Open in a new tab for more thought. Then Mark All as Reviewed. And that is for working from Unknowns for the Rest of Africa. Scrolling down thru a hundred would put me off iNat forever :~)
Thanks for your broad encouragement. I am dutifully dicotting but not convinced that I am getting much response. If I can at least start at family …
Some clear feedback in numbers from the CNC Challenge.
When IDs closed at midnight last night.
Dicots - 29,172 observations with 2,099 species (40% RG)
An awful lot of, blurry green smudge, definitely planty, which is not ever going to move to a better ID.
The reason “Amygdaloideae” don’t improve is that they are really hard to ID! Putting things into this category is a help, but getting it further is just hard.
Annotating phenology, I notice not a lot of observations use this yet to help with IDs. Since flowers are a key feature to be able to sort plants out into families, it could be very helpful though!
If you don’t know much about plants but feel you can recognize when something has a flower, you can help by identifying unknowns or Plantae to Angiospermae and adding plant phenology = flowering. (It won’t work on Plantae since that includes non-flowering plants so the phenology annotation option is not available.)
So far that seems to be a fairly small pool but if we can get more people to add phenology annotations when they see observations with flowers, that might help to shift more of the unknown plants down to families.