Lightly obscure all records on iNaturalist

Feature Requests
1

In my short time on iNaturalist I’ve come across a number of discussions regarding the potential of poaching due to the public nature of iNaturalist locales. I don’t believe poaching is a serious conservation concern for 99.9% of species, but it does happen. The automatic obscuring of sensitive species is supposed to be one way of controlling this, but there is very limited correlation between which species are listed as sensitive and which species are being threatened by poachers. Users can choose to obscure sensitive records on their own, but we know that knowledge an compliance there will never be 100%.

Also, I know in my own case the huge difference between an exact-locale record and an obscured record makes me more reluctant to obscure. I’d like to be able to verify the location of an animal within a particular county, forest, park, etc. without necessary putting someone on its doorstep.

I propose that there be an automatic obscuring of every record on the order of perhaps 2 km. Approved data requests would still get exact data, of course. I feel the downsides would be very minimal - all regular iNaturalist benefits would still be there. And the poaching threat wouldn’t entirely disappear. But it would at least force poachers to work a lot harder. It might not be the perfect solution but it’s one for which I see the benefits over the status quo certainly outweighing the minor downsides.

I’m happy to hear what others think.

2

With the system as it currently works, this would have unintended consequences on smaller-area projects. Not that it’s a necessarily bad idea, but something would have to change in how observations are added to projects before it happened.

I’ve been working on species inventories for a relatively small park, and we’ve found that any obscured observations that have any part of the uncertainty box outside of the project area do not go into the project - a bit of an issue, since the all the endangered species we found wouldn’t show up on our species list!

5 Likes
3

Perhaps a system where users can flag species that should be considered sensitive would be less drastic than obscuring the locations of all observations. In my opinion 2 km isn’t necessarily enough to prevent poaching yet it’s enough to mess up all the data on small scale projects so we’d be losing out on both accounts. If you are finding a lot of observations of organisms you feel are about to be poached a temporary solution could be to either directly message the observer or comment on the observation explaining the importance of obscuring locations on certain species.

5 Likes
4

This has been discussed in length. If it really were implemented it would break Inaturalist for pretty much everything I use it for. I’d probably stop using it.

11 Likes
5

Even better if it’s a species wide issue is to flag the species itself and start a discussion there. Many species are auto obscured for this reason. A user can also choose to obscure their own observations for whatever reason they want.

4 Likes
6

My opinion is that the current system – sensitive species being obscured automatically to 10km, with users then able to obscure any individual observation similarly – is fine.

8 Likes
7

Very true, in both directions.

That system already exists, for exactly this reason. If someone sees a taxon threatened by poaching and not obscured, please flag it for curation. Add comments supporting the existence of a real poaching threat due to knowledge of iNat locations since, as noted, mere rarity does not always equal threats from poaching. (Rarely so, in my experience).

  • Remember that one’s own observations will always appear unobscured (I forget this more often than I should! :man_facepalming:) To verify current taxon geoprivacy status, look on the Status tab of the taxon page.
  • Note that it can sometimes take our volunteer curators a while to notice a flag, so if it’s a really urgent situation, either @ tag a curator that you know in the flag comments, or email iNat staff at the help address.
7 Likes
8

I agree. I love being able to look for a specific species, and knowing the exact best place to look for it, not just the general location. Obviously, this does not outweigh the negatives for a few species, but for a vast majority this is perfect.

2 Likes
9

Most of my observations are in three places/projects. They are all within a 2km area and one is inside one of the others, so even a very small degree of obscuring would make them all useless for my purposes. I’m hoping to get more local people to use iNaturalist and I’d be very interested to see where there are any hotspots for observations, but that would be impossible without accuracy of at least a few metres.

2 Likes
10

This actually came up in a discussion of small projects - both of my projects are already affected by the fact that many species are obscured, often unnecessarily (who could ever perceive a need to obscure Northern Red-legged Frogs in Oregon, for example?).

A solution I proposed in that thread was that sensitive species could be added to a project inventory with the contributor’s permission so long as the project had been in existence for X months (say six?), to prevent poachers from trying to game the system by setting up a bunch of overlapping projects in order to nail down specific observations. If that was in place then projects would be not only unaffected, but would actually be better off than with the status quo.

2 Likes
11

In terms of being able to look at the exact spot where an observation was made in order to view that same animal or population, to me that already sounds troubling even without the poaching threat. Perhaps you are the only person in the world with that particular desire. But as iNaturalist continues to grow in popularity, if any significant number of people actually want to visit exact coordinates for a sighting, that could easily have a detrimental effect on that plant, animal, or fungi whether or not they take it away. I’ve heard that specific complaint in terms of birds (disrupting nesting and other behavior), reptiles and amphibians (disturbance of habitat), and big trees (trampling of habitat), and imagine it could have a detrimental effect on just about any microhabitat.

It also kind of violates the ethic I grew up with in wildlife viewing, which is that the search is the major part of the experience.

If there is a particular species you really wish to look for, it’s always possible to contact the data provider and ask them directly, right? It is when strangers wish to do so without revealing their identity that trouble starts.

3 Likes
12

Can anyone clarify - if one can still contact a data contributor to get the particular coordinates for any particular record, and if well-established projects were still able to list obscured records, what would be the specific downside to this proposal that would outweigh 95% of species being revealed to poachers down to their exact coordinates? I really don’t know what people are talking about exactly, if it’s just desire for extra-easy life-listing or something else I haven’t thought of.

13

I started this topic due to a specific Facebook thread that was concerned with current iNaturalist policies. Here are some of the responses in that thread - nearly all of these are different people:

"I was just researching a state listed species in google when I blundered on to a crap load of Inaturalist records with photos and explicit street addresses.

While I’m real happy to have had my search made so easy, really people? Is that amateurs who don’t know any better? Or is this the high tech evolution of the “Listers” who brag on how many reports they make?

I am definitely talking about a poacher friendly animal for sure."

“This is why I recommend HerpMapper over iNaturalist. HerpMapper by default obscures records to the county/province level”

“There’s been herpetologists and biologists who are trying to contact iNaturalist because it gives poachers exact locations. I agree with using Herpmapper. If you feel strongly about it try contacting iNaturalist. Many have without much response but if more people voice their concerns, maybe they will change how they represent location data.”

“After introducing a bunch of people to iNaturalist I’ve found that for some reason they CANNOT figure out the location settings when posting observations.”

“poaching is a real problem in Maryland . You can’t tell anyone where you find stuff . And I know this sounds crazy but I think someone has been going around killing copperheads in Baltimore County this year”

“Blanding’s turtles, spotted turtles, wood turtles, massasaugas, and even five lined skinks have been hammered hard in Ontario Canada and around the Great Lakes. In one case, 3/4 of an entire wood turtle population was taken in 2 or 3 visits - more than 200.”

“I can tell you overcollection had a drastic impact on the corn snakes in Jasper County, especially in the vicinity of the Okeetee Hunt Club, and that goes back a few decades. Poaching and overcollection has always been a problem. Today, it’s unfortunate that so many herp populations are so badly fragmented, that some collection really takes a toll. That’s definitely the case up here where I live. Texas, Arizona, and some other wild places still have decent populations of herps, but even there, there’s poaching pressure on twin spotted rattlesnakes, for example.”

“I only use iNaturalist for researching new locations to look for protected species. People are dumb, often reliably so. I’m surprised it’s still up and running the way it is.”

“I have to admit, as someone who has been researching this kind of thing since it meant physically visiting college libraries, I can’t get used to google giving me the street address of a listed species in 90 seconds.”

“That’s exactly why iNaturalist needs good IDers. Good observers staying away is one reason the data can be off.”

“I only post common species to iNaturalist. Honestly I mostly use it for fish. I will admit, I kind of wish there was a way to keep off people who are on there with bad intentions.”

“I would say that the majority of those iNat observations that you are looking at are general naturalists or nature aficionados that don’t understand or know the sensitivity of the species that you are looking for info on.”

And finally, this Facebook update was posted into the thread:

“I’ve been MIA posting because it’s been a busy summer this year. This represents only a fraction of the confiscated animals I’ve been dealing with. These Eastern Box Turtles are highly sought after for their incredible colors. US posachers sell them illegally to buyers in Asia, Europe, and even the US. I’m working with some good friends at northeast PARC, local universities, and state and federal agencies to figure out a solution to the madness.”

I myself have found unusual morphs of otherwise common species whose location I obscured because I knew they would be poacher targets. But how many observers don’t know to do that? If we really obscured every poacher-targeted species to 26km or whatever the status quo is, the obscured list would probably increased by a factor of five at least. And how devastating would that be to everyone’s inventory projects? Not to mention how many species in non-Western countries that don’t have advocates telling iNaturalist their records need to be obscured. That’s why I feel some sort of duel system is necessary.

4 Likes
14

Just a note, unless something has changed, it is 0.2 degrees of obscuring, which depending on distance from the equator is about 22km, not 10km.

2 Likes
15

I agree that there are groups that merit a lot of obscuration - many herps, cacti, orchids, carnivorous plants, and I’m sure other folks can fill in more. And plenty of these already are obscured here. If we are missing some, I would respectfully suggest spending this time getting a list to iNat ASAP so we can check into them and plug the holes where needed. (Just be prepared for questions on any species where the need isn’t immediately obvious.)

And there are many other things – most of the 360,000+ vascular plant species, even more invertebrates, most non-vascular plants, and I would guess things like Western Fence Lizards and American Bullfrogs – that have no need to be obscured. They are either not poaching risks, or everyone already knows how to find them without consulting natural history databases. (Think saguaro cactus, giant sequoia, etc.) Or they could be invasive species where knowledge of exact locations can be crucially and immediately important.

Folks use iNat data - in situ or downloaded - for all manner of conservation and research interests. Some of them might not be impacted by your proposal, but many others would be. Often observers will post a few observations and then go inactive on the site, which can leave interesting data points permanently masked.

As one personal example, this observation was very important to me in relocating and further documenting an undescribed species in my floristic area of study. Fortunately this observer was and is still very active, but even if they hadn’t been, the unobscured location data got me to the right spot for further study.

Like @charlie said, global obscuration, even at 2 km, would break the site for many users. And as I mentioned in an earlier post on this thread, it’s not hard to get more things obscured on iNat if you are willing to go through the process and provide supporting information.

And finally, as has been said in other threads, don’t post secrets on iNaturalist, or anywhere else on the Internet for that matter.

4 Likes
16

I’m interested to know what those uses are that would break the site for users. The example you just gave is an extremely unusual occurrence and even then as you note that particular instance wouldn’t have been affected by my proposal. The hypothetical you suggest (an observer posts an undescribed species, leaves it unobscured, and then disappears and refuses to respond to messages/emails with no other contact information or real name, and there is a particular person who was searching for that particular species and can’t find it on their own and needs that exact location and not just the general area in order to find and describe it) has to be a vanishingly rare occurrence, and even if this hypothetical situation does occur once it would just be one opportunity missed, it wouldn’t actually break the site for anyone.

The problem with me recommending more species for obscuring is that doing so WOULD break the site for many people. I have two projects that are already hampered by obscured species. Every time I recommend a new species, it would further hamper my project and everyone else’s, especially due to the currently enormous (~25km) degree of obscuring.

Why is there no middle ground between either making a species ~25km obscuring with no exceptions (which is incompatible with the vast majority of projects) and or making it exact locations across the board unless the contributor is informed enough to know otherwise?

I do feel at quite a disadvantage in this discussion as the large number of people who hold this side of the issue, as I demonstrated, have refused to engage with iNaturalist as a result.

3 Likes
17

Fair enough. I’ll let others continue to chime in on how a universal, unselective middle ground would or would not impact their use. Tracking of invasive species and roadkill issues are two I hear discussed fairly often.

Housekeeping note: relevant ongoing discussion here:
https://forum.inaturalist.org/t/danger-of-locations-on-inaturalist/6602

and in other topics linked therein.

1 Like
18

Note that this is equivalent to a user obscuration of 11km (i.e. diam 22km). But looking at resolution of data for Black Rhinos, the majority of observations have a location accuracy far more vague than the obscuration: most visitors to our national parks dont seem to know where they are - location accuracies (inaccuracies or location error) of 100km and even 1000km are not uncommon!

4 Likes
19

For the present, I think, the obscuring system for threatened species is OK for now, but I noted a problem from a different angle. If a user is submitting several OBs from the same point and one of the OBs is obscured, the (rather) exact coordinates can still be very easily found out following the time line of the other observations from the same point by the same person. By using calendar on the observers profile, for example.

3 Likes
20

There is ongoing discussion about a fairly controversial proposal to deal with this here:
https://forum.inaturalist.org/t/obscuring-geolocation-by-date/6597

3 Likes