I’ve wrestled with this for a long time, probably even before iNat, when I find something that is stationary and could be desirable to some people for whatever reason, poaching/foraging/collecting in places where it isn’t legal. How do I record/document an observation without giving away too much location info for those organisms that aren’t already protected?
It hasn’t mattered that I haven’t posted my fungi observations with regard to that since in places that I frequent the most, as many things have disappeared/gotten cleaned out, or someone picks something and leaves it sitting by the trail (and yes, I know that seeing an entire mushroom is important for ID, but it could be stuck back in its hole when possible).
As I like to photograph things as they are/in situ, that is annoying. To me it’s like finding carved trees, graffiti or trash, or logs rolled and not replaced, etc. out in nature, little respect. Plus, it removes the excitement of finding something yourself.
I usually change the text location that shows up in the map for an observation so that all the observations for a particular park, say, all say the name of the park instead of what Google thinks is an address/county for some, for my own use.
I know that obscuration has changed over the years for sensitive species. And it used to be that those don’t appear within a location-based project, I assume that’s the same way.
But what is the best way to obscure something stationary/uncommon but not protected, while keeping enough info intact for research? I’ve thought about using one set of coordinates for things like that, within said example park, while keeping the exact coordinates to myself.
The best way to obscure is to obscure every photo taken on that day at that location, regardless of rarity. If someone needs the coordinates and you can trust them then you can DM or trust them with hidden coordinates.
You don’t need to obscure the whole day, just those observations, upload them on a different day so people won’t find out where you was. Please don’t upload things with huge accuracy radius, you won’t remember an exact place in years and after your death they will be gone forever, obscuration saves the correct spot, but others don’t see it until you want it.
It’s not clear to me if you are familiar with the geoprivacy functions on iNaturalist. You can select “obscured” for any observation and the public location will be moved to a random point somewhere in a roughly 20 × 20 km box. The true coordinates will be visible only to you and to anyone you choose to share with via projects.
One question to ask yourself is whether you want people to see the observation date. If you want the date obscured, then select “obscured” for the location. But if you want other people to see the date, then you need some workaround like setting the geoprivacy to “open” with a large uncertainty radius. This thread explains how to do that: https://forum.inaturalist.org/t/obscured-observations-only-showing-month-observed-for-other-users/30315
I post lots of moths from my backyard. Originally, I set them to obscured, but I later switched to open with a large radius. Flight period data is very important for moths and I didn’t like the dates being hidden.
I have no idea where you are in the world, but in case you’re in the UK and for the benefit of any UK readers of this thread I would point out that in the UK at least the National Biological Network (who get data exported from iNat) are very clear that they prefer for you to use a circle with larger uncertainty radius instead of the ‘obscured location’ function, because they are not able to export and use the true location of obscured observations. How big that circle is is up to you, but obviously the larger it is, the less useful it is to researchers.
That’s unfortunate, since it isn’t really how the site is supposed to be used. I understand people do it, and it isn’t worthy of trying to make individuals stop, but a major conservation group shouldn’t be recommending it. It’s bad data practice too, and it messes up the maps because when you are looking at a species map on iNat it shows the pin as if it were an unobscured location, and there’s no way to distinguish it from real locations without clicking on each pin. And if given permission the group would be able to export the true location. If not given permission they shouldn’t have the data.
I’m not really sure what you mean. I’m not 100% sure of the details of the data privacy issues and I hope they can find a better resolution, but regardless for now, if the true location is within the circle the data point is accurate, and the uncertainty almost certainly far smaller than for obscured obs. Location data has never traditionally been recorded with exact pins in a map: I’m not sure what it is like elsewhere, but in the UK traditionally records have been located using grid references - which only provides a level of accuracy determined by the size of the grid square (4-figure or 6-figure grids) the only real difference is that that results in an uncertainty square rather than a circle.
Obscured locations are downloaded as a point in a random place in the map within a c.25km uncertainty of the true location, it results in some records being sent to the recorder for the wrong county! I have removed all of my obscured locations and replaced them with uncertainties of <500m - which surely messes up maps a lot less…
You say it as if it’s a good thing, it’s better we can use pinpoints and making huge 500 m circles just to obscure a thing is 1) doesn’t really help hiding it, 2) you lose significant amount of data forever, 3) no problem with allowing researchers to see your hidden coordinates, iNat doesn’t exist to help other websites in that matter.
It should be possible for the National Biological Network to address this via a project. Observers can join the project and choose to trust the project managers with obscured locations. In this way, the finer-resolution data will be available to NBN, and the public-facing data will remain obscured. This would seem a far better solution than asking people to deliberately downgrade the accuracy of the data. I would have thought that anyone who is willing to alter coordinates would be more than happy to simply sign up for a project - it would mean less work for the observer and better data for the NBN.
I guess that depends on whether it is the thing or the place you are trying to hide. (i.e. is the concern privacy or conservation), and to what degree of uncertainty you want to hide it. At the end of the day data belongs to the producers of that data (as you say, iNat exists for iNat users) people can choose whether to make their data more or less usable for whoever.
Anyway, 'tis what it is. Let’s not take the thread off track.
Sure, but we shouldn’t advertise ways that have significant downsides to them, most people who use obscuration of boths types are into hiding place vs. the organism, though I’m not sure what you mean by that, in both cases I would use official obscuration, gladly there’s no reason for me to do it, I have a total of 1 observation manually obscured.
If you look at the individual observation page, or download a CSV with location certainty as a field, the spatial data is accurate (though not precise) but on the species maps on iNat and such, the data is inaccurate, sometimes dramatically so, because there’s no way to designate just by looking at the maps, that the observation has poor location confidence. Ideally any observation with more than ~100 meter accuracy would show the little obscured circle or something, or just not display on those maps, but that isn’t how it works now. I often look at the taxa maps for a species to look for interesting outliers and/or possible misIDs, but it’s common for me to click on one of these ‘outliers’ and find it just was mapped very coarsely. Whereas if it shows the obscured symbol i won’t click on it unless it is way out of range.
When you download obscured observations, if you have permission to see the true locations, they are downloaded in a separate column. What’s annoying is it only fills in the true locations of the obscured observations, not the open ones, so if you have a data set with both you have to manually move over the open observations’ location, which is a pain especially with large datasets. But it is doable. I’ve done downloads of my observations with true location including my obscured ones, that i use for personal/work use.
I’m sorry, I don’t understand what is hard in realizing that you can just check previous and next observation uploaded and find out the exact date, time and place? You must upload obscured observation on a different day to actually obscure them.