Differences in id quality and speed

General
1

We live in central Tanzania, working in a project for regenerative agriculture, and I have uploaded observations to inaturalist für a couple of years now.

Some things get id’d pretty accurately by the geomodell, or by other members. But other things take long or never get id’d.

Observations that get an id quickly are mainly mammals and birds. Observations that are more difficult are insects and plants. Those often get only to the point of genus or family.

I have a hunch on why that is: plants and insects are incredibly diverse, and also there are much fewer experts on African plants and insects than in other parts of the planet.

Am I correct in my conclusion? And is there anything that could be done to improve this situation? I would really like to learn more about all these wonderful parts of God’s creation around me, but I often get stuck…

Thanks in advance for any helpful thoughts.

10 Likes
2

Yes, but it’s also very difficult to ID some species from pictures alone. Some plants require pictures of the flowers, so it’s also about timing.
For insects, sometimes wing venation can be important, or the hairs or spines on the legs, or some other bit of anatomy that you may not be aware of, or is specific to that group. So you need very clear very close photos, and probably some familiarity with that group.
And that’s if they don’t require dissection! It can be frustrating at first, but manage your expectations for insects and it becomes more fun : )

I don’t know about your knowledge of insects, but if you yourself are able to ID to family level for example, it can get to the experts of that group more quickly; many entemologists are forced to have a narrower niche than an ornithologist or similar : )

11 Likes
3

Yes, your observations are correct, and @apusaffinis has expanded upon them very helpfully. It is not just where you are - insects and plants are identified more slowly than birds and mammals in all parts of the world because, as you say, they are more diverse and there are fewer experts. It is also true as @apusaffinis has pointed out that it is the very nature of the beasts that means they are harder to identify - and often impossible.

Your location exacerbates this because the Afrotropical fauna are simply not as well studied as the Palearctic or Nearctic fauna for example. This is one reason why these observations are so important. They may not be identifiable - yet - it may be years before science catches up with your observations. Your region is full of insects that are undescribed, or just plain unknown to science, or where there is confusion about where the distinction between species lies, and how they are to be identified. The Oriental/Indo-malayan region is similar. There are many species on iNaturalist in these regions which have had to be identified by reference to obscure and often very old scientific papers because they have never been photographed before - and this is quite a challenge.

By providing these observations you are generating the raw materials for future science. Which is awesome. Don’t be discouraged because of a lack of IDs on them :)

19 Likes
4

Echoing what others have said. These taxa (plants and insects) can be tougher in general, and in your region, the issues are also higher diversity and relatively fewer experts. So what is in your control that you can do?

  1. Take high quality observations by photographing as many key characters as you can. Multiple photos showing different parts of plants (whole plant, leaves, stems, flowers and fruits if they exist). For insects, I just try to get a lot of different angles, as I don’t necessarily know what is important.

  2. Get to know expert IDers in your region and tag them judiciously and ask for advice or resources that can be useful in making the best initial IDs - this will get your observations to those that can ID them (if those experts exist!) more quickly.

I agree with @matthewvosper - don’t get discouraged if these are not IDed right away or at all! If you are making good quality observations, they are likely to be useful to someone at some point in the future who may be able to ID them. I think most folks that have been on iNat for a longer period have the experience of an observation that is years old suddenly being IDed out of the blue when they thought it would never be IDed - a nice surprise!

8 Likes
5

If. You can start your ID at family - then the taxon specialists will filter for it.

In August we did Mission Impossible - ID plants in Africa. ALL the Unknowns were cleared (but for some that meant moving them to a still broad ID like Dicots)

Flora of Africa - despite the given challenges - is two thirds at RG. Which amazes me, and makes me happy!

For insects there are active specialists for butterflies, bees, flies, mantis, orthoptera …

6 Likes
6

86% of all bird sightings on iNat are research grade (around 23 million) while 61% of the 60 million verifiable flowering plants are research grade. As someone who posts mostly insects and plants, 43.5% of my verifiable observations are research grade.

Even in North America, there are many regions and taxa lacking identifiers. Too many users, too many species across iNat. When I’ve tried to help with African wasps, best I can do is family or subfamily and I rarely see follow-up notifications outside of South Africa.

6 Likes
7

Two more things that are within OP’s power:
1 do some identifying (if they don’t already). As well as being helpful to other users, trying to identify other people’s observations provides great insight into what sort of photos are most useful. They don’t have to be experts in anything, they can work on categorising unknowns into high-level taxa, or they can try building expertise in something.
2 Encourage other users in the region to also do some identifying. iNat has a shortage of identifiers in general, and talking people into helping is generally a good idea.

6 Likes
8

Not getting ID’s is frustrating, but please keep on posting. Your observations are especially important because of the lack of observations in your area. Please keep going.

5 Likes
9

For example
@chemp
who I call in for African Orthoptera.

If you work thru IDing for Africa you will notice - new active identifiers.

Pumpkin family
@hanno_schaefer
@norbert_holstein

But the usual rules about @mentioning. Not too often. Good clear pictures. Try your best to ID as far as you can. Please respond to notifications.

We have found some very rewarding new species. You can check the monthly CV updates for your taxon and location of choice.
Using data from 20 August
https://www.inaturalist.org/blog/84723-a-new-computer-vision-model-v2-7-including-1-785-new-taxa

2 Likes
10

FWIW, only 28.5% of fungi worldwide are research grade. Some taxa have fewer morphological features that work well with photo IDs, fewer experts to identify them, etc.

4 Likes
11

Anyone who wants help with Ants, Vespine wasps, or, if in North America, any Hymenoptera, can ask me, I like identifying these and would like to reduce the difficulty people have getting IDs on them (though I am not super familiar with African species)

Species level ant IDs are really, really, hard, you may need specialized camera equipment at just the right angle with the lighting just right, for example this ant I posted is just not IDable to species even with my specialized equipment), you would have to analyze the relative lengths of the segments of the mouthparts https://www.inaturalist.org/observations/183405147

And this one is only identifiable to species by the fact that it has upright hairs on the antenna, and I was able to zoom way in since it stopped moving to eat https://www.inaturalist.org/observations/183751329

But not all are like this, for example this is IDable to species despite me failing at steadying the (non-specialized iphone) camera https://www.inaturalist.org/observations/178799915

Don’t let a lack of special equipment discourage you however, I was able to take this with just an iphone when I didn’t have my DSLR & macro lens with me https://www.inaturalist.org/observations/178452633

For insect ID you want to have some clear up close pictures (or as close as your equipment and risk of stings allow), don’t only rely on zoomed out shots of entire nests, a good rule of thumb is that for ants and other small insects 1 mm is equivalent to 1/3 of a meter for normal animals, so taking a picture with the camera a meter away is like trying to ID normal animals from 300 meter high aerial photography

And I know I keep saying you want to zoom in as close as possible, but don’t do anything dangerous, there are some things you just shouldn’t ever get close to, in my area (northeastern US) those things (insect wise) are nests of the genera Vespula and Dolichovespula

EDIT: Speaking of dangerous stuff, are the larger Dorylus species (driver ants) actually deadly like people here think they are?

6 Likes
12

Funny thing about Hymenoptera. This is identifiable to species https://www.inaturalist.org/observations/162030508

image
image1614×1615 119 KB

and this one I can’t even find the genus name https://www.inaturalist.org/observations/141322151
image
image1920×1920 147 KB

8 Likes
13

Thats an awesome in flight photo!

3 Likes
14

Very understandable. In most cases, a botanist needs a dried sample to investigate different morphological features. I suggest that you start doing your own collection or small hebarium then visit Science universities in Tanzania, get to know Tanzanian botanists and ask for identification. Botanists are happy to receive samples and will add it to the herbarium of the university as a gift from you. This way you will have contributed to the science in this outstanding country while you obtain a trustworthy identification.
As for the insects, the same principle goes if possible. If not then take as many images as possible from different angles. There is a possibility that you are witnessing species that are new to science.
An important note is not to over collect samples. Perhaps the plant you are taking is so rare or the insect as well. Lots of luck to you.

1 Like
15

How are you with Caribbean species?

16

Better than Africa but not great once you get south of Florida and Texas

17

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.