iNaturalist

General discussion on entomology
User avatar
58chevy
Meek
Meek
Reactions:
Posts: 375
Joined: Mon May 23, 2022 5:58 pm
Location: Houston, TX USA
United States of America

iNaturalist

Post by 58chevy »

interesting short paper published yesterday in PLOS Biology encouraging experts to contribute identifications to iNaturalist.

----------------------------

In recent decades, there has been a massive increase in available biodiversity data—there are currently >2.1 billion species occurrence records in the Global Biodiversity Information Facility, representing a 12-fold increase since 2007 [2]. This rise in biodiversity data is due in part to the growing popularity of citizen, or community-based, science.

One of the most globally successful platforms is iNaturalist (www.inaturalist.org; [3])—a multitaxa platform and joint initiative of the California Academy of Sciences and the National Geographic Society. <snip> While the quantity of data and contributors continue to increase on iNaturalist, one bottleneck to fully realizing the potential of these data for scientific research is the dearth of participants with reasonable expertise (i.e., someone with the skills and ability to make informed identifications)—hereafter “identifiers”—actively participating in the community. The iNaturalist community—as of January 2022—consists of 2.5 million users, 92% of whom only observe, <1% of whom have only made identifications, and 7% of whom both observe and identify. More recruitment of identifiers is clearly needed. Here, we provide our collective perspective on 7 reasons to contribute to iNaturalist as an identifier (Box 1).

Callaghan CT, Mesaglio T, Ascher JS, Brooks TM, Cabras AA, Chandler M, et al. (2022) The benefits of contributing to the citizen science platform iNaturalist as an identifier. PLoS Biol 20(11): e3001843. https://doi.org/10.1371/journal.pbio.3001843

---------------------------

Mike Quinn, Austin
User avatar
adamcotton
Global Moderators
Global Moderators
Reactions:
Posts: 734
Joined: Tue Mar 22, 2022 12:24 pm
Location: Thailand
Thailand

Re: iNaturalist

Post by adamcotton »

Unfortunately, due to the 'majority vote is correct' policy, many experts do not want to contribute because they see it as a waste of time if their opinion is 'voted' down by a load of people who don't know enough about the taxa involved.

Adam.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

adamcotton wrote: Sat Nov 12, 2022 12:10 pm Unfortunately, due to the 'majority vote is correct' policy, many experts do not want to contribute because they see it as a waste of time if their opinion is 'voted' down by a load of people who don't know enough about the taxa involved.

Adam.
I think that's in line with the appeal for identifiers in the paper.

The risk isn't so much that expert IDs get voted down, it's a race to "Research Grade"; once there are three IDs at the species level an observation becomes "Research Grade" and is considered fact. The problem is that yes, there are a lot of morons who "identify" observations in order to get high ranks as "Identifiers" and achieve some sort of notoriety.

One can dispute the taxon. I have. At that point the non-specialists (AKA birders) usually step aside and let the pros settle it. Of course there are problem Identifiers who take offense at being corrected- that happens in the print publication world too.

As we've previously discussed, there ARE a lot of faults with iNaturalist. I will not rehash them.

But what iNaturalist provides- and does so marvelously- are the observations by generalists that would not otherwise come to the light of day. AND, the ability to quickly view high-level data in an easy to absorb format. For example on the former: At a whim I searched up the Eyed Elater (Alaus oculatus) a species which I've encountered in my area three times in my life- all chance encounters. Sure enough, there's a bunch of observations in my area, chance encounters by non-experts with cell phones. Bravo. Example of the latter: want to map Papilo glaucus dark females with the Battus philenor they purportedly mimic, and philenor's food plant? Mostly easy- the maps for Aristolochia and philenor readily demonstrate a substantial range overlap; the black female glaucus has to be mined image by image, but that took only a couple hours. You won't get that from any publication.

I did voice a question on iNaturalist whether non-expert observations should be throttled, and as one would expect I got slaughtered by the "everyone's opinion is equal!" crowd. I understand we would not want to prohibit sharing of observations, since observations are what makes iNaturalist so valuable. And, sure I'd like to ban some Observers, just like I'd like to ban some people at work from opening their traps, and like to ban some constantly-wrong morons from internet forums...not going to happen.

So the solution is indeed to get experts to vet that data (observations) and that's where experts are needed. Maybe iNaturalist needs to have a "blue ribbon" for experts in various categories, I don't know. Already there is a lot of garbage data, and already it's too much to clean up. But if experts were to start Identifying then perhaps even more garbage data wouldn't get through. So again, more experts needed to lead.
User avatar
Cabintom
Meek
Meek
Reactions:
Posts: 257
Joined: Sat Jun 04, 2022 1:54 pm
Location: D.R.Congo
Canada

Re: iNaturalist

Post by Cabintom »

I'm the most active identifier of Sub-Saharan butterflies on iNaturalist & probably among the top 5 butterfly observers north of South Africa (I haven't checked, so I'm not sure where I "rank.")
adamcotton wrote: Sat Nov 12, 2022 12:10 pm Unfortunately, due to the 'majority vote is correct' policy, many experts do not want to contribute because they see it as a waste of time if their opinion is 'voted' down by a load of people who don't know enough about the taxa involved.
The reality is that outside of the more developed countries there simply aren't "a load of people" identifying butterflies. From experience, I'm usually the only person, outside of the observer, providing IDs for most Central African & West African butterfly observations. Ok, there are the "habitual confirmers" who just agree with "expert" IDs, but they're not all too common and it's at the point now where if I provide a contrary ID there's often no discussion and they follow suit. Is that a good thing? I don't know. But for most people, good reference material for IDing is inaccessible.

South African observations on the other hand usually have several IDs. When I've disagreed with the established majority, almost all of the time it's been easy to convince the others to adopt a corrected ID. But, I have established myself now as an expert (I guess) & most "power users"/curators respect my opinion.

My point is that, at least for areas which have few identifiers, once an expert's expertise is recognized (a good profile helps), then they have more influence & the perceived challenges of the system are minimized.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

Tom, I don't know where you get the time to do all that you do!

I read that 50% of butterfly observations on iNaturalist are monarchs. That's nice. Meanwhile, I have several observations of known species (including Papilio canadensis x glaucus) that are only at 2/3 so not "research grade." I guess it's easier to rubber stamp common stuff.

And as I'd mentioned the "Identifier" collectors. I posted an observation of a fish (genus Esox) that appeared to be a species not known from that body of water, hoping to get a real expert ID. Obviously it's Esox. Yet some dumbass "Identifier Collector" who has 21,500 Identified adds one more with my observation, identifying it only to the genus Esox. Well duh, Mr. purported expert, that was real useful.

I do wish iNat was used more for collaboration. There's the ability to add comments to observations and identifications, no reason why there can't be some back-and-forth. It's possible that iNat will mature to that point as the crowd-sourcing youth participate; or, it could go the same way as forums with the dumbing down and just posting pictures.

It's great that Tom spends the time to ID the African species; over time this will become invaluable. Wish we had expert Identifiers for other regions/ taxa. Why not get on over there and help, everyone?
User avatar
wollastoni
Site Admin
Site Admin
Reactions:
Posts: 439
Joined: Fri Mar 18, 2022 9:51 am
Location: France
France

Re: iNaturalist

Post by wollastoni »

Chuck wrote: Mon Nov 14, 2022 1:14 pm Why not get on over there and help, everyone?
I have just registered and found some rare live pictures of scarce Papuan Delias. I have asked their authors if we can add these live shots on our Delias website (with full credit to them of course).
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

wollastoni wrote: Mon Nov 14, 2022 4:58 pm

I have just registered and found some rare live pictures of scarce Papuan Delias. I have asked their authors if we can add these live shots on our Delias website (with full credit to them of course).
Sometimes some observations don't get identified to genus, so you may also search the specific area for all butterflies. Note: I just did this for New Britain and New Island, no Delias.

However, case in point, there's a record for Delias callista that the observer originally posted as Papilio (and it was later corrected) so when first posted it would not have shown up in a search for Delias. Another case supporting the appeal for more expert Identifiers. In many cases in developing countries, it's not experts on any particular class that takes the photos, it's someone with some volunteer/ Non-Governmental organization- and they really do get to remote places, and often for extended periods.
User avatar
Cabintom
Meek
Meek
Reactions:
Posts: 257
Joined: Sat Jun 04, 2022 1:54 pm
Location: D.R.Congo
Canada

Re: iNaturalist

Post by Cabintom »

Chuck wrote: Mon Nov 14, 2022 7:07 pm Another case supporting the appeal for more expert Identifiers. In many cases in developing countries, it's not experts on any particular class that takes the photos, it's someone with some volunteer/ Non-Governmental organization- and they really do get to remote places, and often for extended periods.
Yes. Exactly. For DRC, butterfly observations are almost entirely coming from folks working with the parks.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

For those who haven't used iNaturalist, here are some top uses:

1. Determine where best to find a species: far better than the old Golden Nature Guide range maps, iNat range maps use actual observations. At the national level, it's easy to see where a species is found, or not often found; you can also zoom in even to the town level to find specific locations where a species has been observed.

2. Determine WHEN a species can be found, and in what life cycle: iNat has curves that show what life cycle a species is at, and this can be refined down to local levels. You can also sort images by date and it becomes pretty obvious from the images the stage in development (e.g., when all the adults have died, you'll see observations of larvae.)

3. Mapping foodplants. Particularly for species that have restrictive preferences for food sources, one can see the range/ quantity of foodplants; no foodplant, probably no populations.

4. Mapping topology: similar to above. However, I use it to search for where one MIGHT find a species that isn't known, or isn't common, simply because records become fewer as the distance from urban areas increase. For example, one might find Species X on a plateau near a city, but also note there's a further plateau that may be less developed- and thus a better chance of finding Species X.

5. Travelling- what's there? My travels take me to places unfamiliar to me. In general, I can search up what one might find well before I get there. For example, I can search for Papilio in Denver in July. What I see from iNat guides me on where to be and what to look for; almost like having local knowledge. Similarly, it can avoid wasting time looking for something that isn't actually there/ then.

6. Identifying plants: It's easy to ID what flower is the nectaring source. I now use it also to ID common plants that I find of interest; previously (as in, for decades) if I couldn't eat it, I wouldn't bother to look it up. iNaturalist makes it very easy to ID plants, if nothing else because I wonder what they are. But it's also been invaluable for recording the plants, sometimes obscure, on which I find Leps; no more "nectaring on purple flower."


The above are all from a user/ benefit perspective. I am a bad Observer- I don't have time to post all of my observations. What observations I do post are those that will probably be of interest to someone: remote locations, species range records (i.e., species I know don't belong there) and I'm working to add Observations of lesser-interest species (e.g. those grey & brown moths). The latter is pretty easy- I don't retain specimens of all the small moths that come to the sheet, though some are pretty interesting; it's very easy to take a photo and upload to iNaturalist.

SEEK is a phone app which is tied to iNaturalist (though iNaturalist has a separate app.) With SEEK you point your phone camera and it will tell you, to varying levels, what a thing is. Noting #6 above, SEEK is the app I use to ID plants. Most of the time SEEK can at least get you to the genus level. SEEK isn't perfect; it said our cat was a fox. But, it's astonishing what it can ID, plant or animal. And, then you can upload the observation to iNaturalist with one button. If you don't have SEEK, get it.

For good and bad, records (Observations) on iNaturalist aren't well vetted. But the big plus is that they are there! In pre-iNaturalist days I caught an invasive fish and flagged it to the state Conservation department - they wanted photos- well forget it, I know WTF kind of fish it is. And, another anecdote, when I noted that I had tons of state and county records that weren't on butterfliesandmoths.org that said I had to send them photos (of species like cecropia, quniquemaculata, etc.) to get in the database- forget it, not worth my time. With iNaturalist, bang it's uploaded. No, not all data is correct, but mine is, and it's now publicly available.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

Interesting and fruitful anecdote about iNaturalist:

Yesterday I uploaded about 60 observations of small moths to iNaturalist.

I was able to batch process the location (since they were all from the same location) but couldn't figure out how to batch process the taxonomy, so uploaded them as "unknown" even to the kingdom. I planned to go back through and ID as soon as possible.

Within minutes, many were being assigned at least to Lepidoptera. Most of this appears to have been done by one of those "Identifier collectors" who get excitement by accumulating identifications. However, I believe this worked in my favor, because through the remainder of the day many were picked up and IDed at least to genus by Lep people.

So now I have a bunch of IDs with zero research on my part.
User avatar
adamcotton
Global Moderators
Global Moderators
Reactions:
Posts: 734
Joined: Tue Mar 22, 2022 12:24 pm
Location: Thailand
Thailand

Re: iNaturalist

Post by adamcotton »

It will be interesting to leave the identifiers to it for a few days and then check the IDs to see how many are correct, how many understandably incorrect (closely related species) and how many completely erroneous.

Adam.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

adamcotton wrote: Thu Nov 17, 2022 9:37 pm It will be interesting to leave the identifiers to it for a few days and then check the IDs to see how many are correct, how many understandably incorrect (closely related species) and how many completely erroneous.

Adam.
You're getting me depressed. But I suppose it's one more experiment.

Perhaps a better one would be to post a Limenitis and call it a Monarch, and see if the idiots beat the better identifiers...could we get archippus to be a monarch?
africaone
Junior Member
Junior Member
Reactions:
Posts: 40
Joined: Mon May 23, 2022 9:24 am
Belgium

Re: iNaturalist

Post by africaone »

I use this site sometimes for my groups. It is true that nomenclature is not at the top despite being at good level. Many of the corrections I made (quite all) were followed and sometimes these corrections have been made in the database following my remarks. It is something dynanmic.
What is amazing is to see sometimes things extremely uncommon in collection with infos on the biotopes and picture in the wild.
The spirit is more to post live specimens and not from collection.
Some people used your experience to increase their own and this is a good thing.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

adamcotton wrote: Thu Nov 17, 2022 9:37 pm It will be interesting to leave the identifiers to it for a few days and then check the IDs to see how many are correct, how many understandably incorrect (closely related species) and how many completely erroneous.

Adam.
Four days later, all ID to genus, about 50% to species. The species level ID are clearly "off the top of the head" (meaning, in US vernacular, somebody happens to know the species), it appears nobody did any research. Two primary species level Identifiers, one Jason at Cornell - so kinda cheating.

I see none that are incorrect.

The remainder I know many are rather common and well known; some are what I'd call "focus groups", genera that have a niche following, and had been discussed on the archived InsectNet forum. So they are known, but apparently the amateur specialists are not on iNaturalist.
ColoradEO
Junior Member
Junior Member
Reactions:
Posts: 5
Joined: Mon Aug 29, 2022 6:59 pm
United States of America

Re: iNaturalist

Post by ColoradEO »

I had 5 minutes so I just ID’d one of yours…. Hope I got it right. Eric
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

Ah, the good and bad. Or, some might say, the honeymoon is over.

I've found iNaturalist invaluable for research, even if some observations have to be manually filtered as junk. I've obtained data, as recently as yesterday, that could not be obtained elsewhere.

That said, there's a dark side.

An inordinate quantity of high-value contributors (e.g., 40,000 observations for one contributor alone) have apparently up and left, and took their observations and identifications with them. Apparently, when a user deletes an account it eliminates all of the observations, identifications, images, everything.

The impact of substantial loss of data spreads across all aspects of research. Note too, publications that reference iNaturalist data thus can incur what is effectively a loss of any value in that reference.

The technical solution, divorcing users from recorded data, apparently has some legal issues (as I understand it.)

It appears, at least to me, that iNat staff is less concerned about what drives critical contributors to leave than studying why anyone might leave (i.e., no differentiation between two groups) and promoting "inclusiveness" (potentially to an extremist level) over data retention and accumulation.

I'm not sure what the outcome will be. It merits watching to see who else leaves with all their data. There may be a lesson here for all crowd-source sites and forums that pushing contributors too far can result in a significant loss (and if it's not clear, with that comes loss of revenue.)
User avatar
adamcotton
Global Moderators
Global Moderators
Reactions:
Posts: 734
Joined: Tue Mar 22, 2022 12:24 pm
Location: Thailand
Thailand

Re: iNaturalist

Post by adamcotton »

That is sad to hear. I hope it doesn't happen often, and a fix can be made to preserve the data.

Several months ago I assisted in peer review of a new paper on Butterflies of Cambodia, which included many new records from iNaturalist. At least these records have now been converted into part of the scientific record.

The Cambodian paper is available here:
https://www.fauna-flora.org/publication ... l-history/

Adam.
Chuck
Premium Member - 2024
Premium Member - 2024
Reactions:
Posts: 876
Joined: Mon May 23, 2022 2:30 pm
Solomon Islands

Re: iNaturalist

Post by Chuck »

Roger that, Adam. In the recent LepSoc, Wolfe "...Siproeta superba euoe..." found images (observations) on iNat, reached out to the observer, obtained permission to use the images, and published them within the paper. This is probably the best way to ensure references are preserved.
User avatar
Jshuey
Global Moderators
Global Moderators
Reactions:
Posts: 148
Joined: Tue Mar 22, 2022 2:27 pm
United States of America

Re: iNaturalist

Post by Jshuey »

An inordinate quantity of high-value contributors (e.g., 40,000 observations for one contributor alone) have apparently up and left, and took their observations and identifications with them. Apparently, when a user deletes an account it eliminates all of the observations, identifications, images, everything.
Wow - who knew that? I'm in the process right now of using the "Wayback machine" (web.archive.org) for 2020, to see if it digs deep into the data set, or just skims the surface. Man it's slow loading on this platform.

After about 5 minutes of spinning - I'm guessing that it can't find the actual data I asked for. So that's not the solution... .

I guess the solution is to download the data (that you trust) on a regular basis as an excel file...

John
User avatar
Cabintom
Meek
Meek
Reactions:
Posts: 257
Joined: Sat Jun 04, 2022 1:54 pm
Location: D.R.Congo
Canada

Re: iNaturalist

Post by Cabintom »

Does GBIF hold onto the data even after it's been pulled from iNat?
Post Reply

Create an account or sign in to join the discussion

You need to be a member in order to post a reply

Create an account

Not a member? register to join our community
Members can start their own topics & subscribe to topics
It’s free and only takes a minute

Register

Sign in