Observations with (automatically) obscured locations are extremely valuable records but generally can’t be used (eg by a Local Environmental Record Centre) as the location resolution is too imprecise. It says precise location is available to someone the observer trusts. How can this be arranged? The particular issue I have in mind is records of Skates (Rajidae) where the records are mainly washed up egg cases, where obscuring the case location seems particularly unnecessary!
1 Like
They could always show their credentials and directly ask the observer if he accepts to communicate?
Although in this case, my guess is that the sea would probably move the eggs kilometers away from the fish, and the next wave might well take them even further, before the conservation center even learns about it on iNat… meaning that more precise coordinates might not really matter?
That said I’m also curious about who exactly has access to the real data for obscured observations, is there a list somewhere with all these scientists/groups, etc.?
1 Like
As I understand it, the unobscured data are available (1) for the institutions running the various iNaturalist communities (see https://www.inaturalist.org/sites/network). But they get it as downloads once or twice a year.
The other way is to set up projects (2) where the users trust the project curators - in South Africa we use this to update the Red Data Listings (https://www.inaturalist.org/projects/south-african-red-list-plants-and-animals) which contains almost all the obscured species in the country (which are maintained by SANBI as a “sensitive species listing” - only species threatened by poaching and harvesting are obscured, not other red list species). But it is a pain, as lots of data are collected by people not involved in the national project (622 people have joined, but 14,711 people have contributed data - how do you chase 14k people to “trust” the curators to access the data?), or dont even know that they have seen an “obscured species”!
Scientists and conservation agencies (3a) can get the unobscured data via the network coordinators (or directly from iNaturalist staff (3b) where this does not conflict with country laws). That is far more efficient than creating a project to trust (3c), or getting personal 1-on-1 trust (3d), especially as users seldom have a way of distinguishing genuine scientists/managers from outright poachers. But my experience is that reserve managers and local conservation units are unable to access the data, and for smaller nature reserves and conservation areas are never alerted to the fact that they have these records in their domains, and often remain ignorant that these species have been discovered in their reserves.
Another major nuisance are people obscuring observations unnecessarily with the aim to providing further security to “sensitive species”. This hides the data from local agencies, and is a pain to track down by researchers/conservationists working the “open” species. In effect, most of this data is lost, unless avenues 3a or 3b are used to access the data, which is usually only viable for a major research or management project.
(a disconcertedly large number of researchers dont realize that the data are “obscured” when they download the data from iNaturalist or GBIF).
4 Likes
iNat’s Help Page on Geoprivacy:
provides some relevant answers.
2 Likes
@tonyrebelo ‘s comments address the main options and their limitations, so I’ll just add a couple of points that may be relevant from a curatorial standpoint.
First, iNat doesn’t currently have a way to apply taxon obscuration differently for different life stages, so there’s no automatic way to obscure locations for adult skates but not their egg cases.
Second, there are many species that have automatically obscured locations by default simply because they have a conservation status of “Near Threatened” or worse, and no-one has taken the time to assess the actual pros and cons of opening up precise location visibility.
The solution to this is for someone to take on that work for a particular taxon area and/or geography. You would need to read the relevant literature and connect with conservation authorities or experts in your area. If you’re lucky, there will already be criteria for or lists of species that are at sufficient risk from poaching that their locations need to be protected.
The logic here really is to consider if there is a realistic threat to the species that is exacerbated by having precise location data. So, knowing the precise location of a rare orchid is almost always a threat, because these plants are coveted and having location info makes them pretty easy to locate. This may not be the case for threatened seafish as they’re very mobile and even knowing the location of an observation on the same day isn’t likely to make them easier to catch. If poaching (or a small number of other risks) is not a big concern, then sharing precise locations tends to be better for conservation.
You can apply conservation status at a global, national or subnational level, so if you establish that a particular skate or ray doesn’t need to be obscured in the UK, you can make the change just for that country. You’ll need curator privileges to do that. If you’re not a curator, you can raise a flag on the species, provide details of your research and ask someone to do it for you. If you’re going to do this for a few species, you might want to request curator privileges specifically.
One possible issue for marine species is deciding what iNat place to apply the conservation status to. Fortunately, there is an iNat place named “United Kingdom EEZ” that seems like it could be appropriate.
2 Likes
This isn’t entirely true. They’d just need to establish a parallel real world network where the true location of each observation is shared with in-network participants outside of inat. Or… the local environmental org… could be the keeper of locations thereby protecting the privacy of participants uncomfortable sharing their real-world location more broadly.
If we’re supporting local solutions, the exact location needn’t be shared globally.
I believe observers are mostly protecting a private place when they chose to obscure its true location…
I have been contacted a couple of times about this geoprivacy problem by other users, sometime complaining about it, but they never asked me to join/trust a project…
So I would reiterate my advice, if you are a legitimate conservation group or researcher, meaning you would either have a website or published papers to show to the observer, you could offer him some incentives to participate in your effort, maybe an invitation, membership to the “club”, in exchange of their “trust” or private communication/discussion about the locations.
This would eliminate potential poachers - even if they were scientists - since the observer could see what they are doing exactly.
1 Like
Oh, I recon I could whip up a few AI papers to show if I was a poacher, and even claim to be associated with a website or two.
But if I was interested in Elephants in southern African I would need to get 14,554 permissions with incentives like membership or other whatever. I “believe” that this would a rather onerous task.
Having helped coordinate several projects - including the South African Red List - I dont think this would work for any serious project. Tens of thousands of users is not really that easy to manage, no matter what one believes.
1 Like
This needs to be considered bc some folks post observations from around where they live, and that’s not something many people want to broadcast. Cybersecurity is a legit concern. We lose observations entirely if we force users to give up privacy for specificity of location info. Also, I never am using geolocation when uploading so it’s always approximate anyway.
1 Like
Yes, liars, cheaters and criminals will always exist (and AI just makes it worse), but this is a totally different issue.
Again, elephants can easily move a dozen kilometers in a few days, so exact coordinates would not be useful. I mentioned this to the OP, fish eggs can be dispersed over 100’s of km, so a precise location doesn’t hold much significance.
I think this geoprivacy option is a good thing, it allows users to keep high precision records while not giving away their address for example, and on the map it clearly shows that this is not the exact place. I find this more practical than choosing a different location, then increasing the radius, it’s also more “honest” in some way…
Another way to do it is to enter the name of a close city for example…and you end up with all kind of wildlife observations in the middle of a busy neighborhood! At least in this case it’s obvious by looking at the map, that these are probably not true coordinates.
On the other hand I believe a smaller grid could be useful sometimes, say 2 km in addition to ~20 km… this is particularly the case when an observation is near a national or geographical border, and ends on the wrong side… but you can’t have it all?
1 Like
One of the species I obscure is Leichardts grasshopper which is a spectacular orange and blue. It is something that some will collect being very rare and unusual. It’s actually not that hard to find if you look at the host plant and work out the type of terrain it grows on and the plant is distinct as well. But we have had poachers in this part of the world so I am not going to make it easy for them.
Past requests for different or user-settable levels of obscuration of coordinates have been denied. Users can achieve this manually; large accuracy values are accepted by the iNaturalist platform and community.
2 Likes