In my case, it helps to find observations that need a fix (for instance this observation whose location I have just fixed, and I fixed the DQA flag for inaccurate location).
But in your case @graysquirrel nothing to fix in the results, almost only observations marked as “it’s as good as it can be” in the results.
I have cleared mine. I don’t have GPS so have to find the ‘right place’ on the map. If I later realise I was wrong … and shovel it around … I seem to lose the Accuracy.
But 3.5K looks like a bigger problem. Maybe someone else can help see where the problem might be?
I sorted mine by date observed ascending - more problems when I was new here.
But I still think iNat should make Location Accuracy Not Recorded more immediately visible to us. I cannot even see it on my own 28, unless I know to keep clicking …
Recently someone identifying my Koelreuteria observations asked me to fix the locations (yet the locations were correct, as far as I could check) and to fix the location accuracy.
I have no time for that!
For most of my obsevations, the location (whether it is correct or wrong) is the one recorded by the camera, and that’s it.
I have 44,965 observations without location accuracy recorded, and 17,536 with it. The only ones that have it are the ones where I manually entered the location and expanded the circle (usually because it’s an old photo with no gps info, and I can’t remember the exact spot it was taken).
I never saw it as an issue tbh - I read “accuracy” more as “uncertainty radius” - and if it’s an exact GPS location there’s no uncertainty, so no need for it.
You must have some very sophisticated GPS equipment if you’re able to log locations with no uncertainty! For one thing, if you’re photographing anything that you’re not standing directly over, the precise location of your camera is not the same as the precise location of the organism. For another, a GPS-lock is not precise to the millimeter.
I input most of my observation locations completely manually, but sometimes the photos from my TG-6 will have location data included. While that’s very helpful for recalling where exactly I took those photos, it’s not good enough to leave as-is. In cases where I do know the precise location that I was standing when I took the photo, I find that the accuracy circle from the GPS pin has to be at least 15 meters (sometimes 30 or 50) in diameter to include it. I’m sure some devices are better than that, but I wouldn’t expect it of all of them.
The reason to include an accuracy circle is so that other people know how precise your locations are. You might know that your phone is always accurate to 5 meters, and not bother to include that information because it’s close enough to the true location that it doesn’t matter, but other people viewing your observation don’t know that. It could be 200 meters from the true location for all they know.
I georeference all my images automatically against a track recorded on a GPS unit that I check regularly and trust to have an accuracy of at least 10 m. I somehow thought the accuracy was recorded automatically (by magic ?) and, like @graysquirrel, I had never given it much thought. I now realise to my horror that NONE of my observations have the accuracy recorded . I’ll try and do better in future, but I imagine there is no way of going back and bulk updating more than 7,500 observations, so as a workround, I’ll be adding a line to my profile explaining the process and specifying the accuracy. Not perfect, but I can’t see any other way.
Cannot find it - but there is a forum thread. 4 decimal places for GPS is enough - otherwise we are saying ‘which side of the snake’ we are looking at. Another layer of the compromise between citizen scientists engaging with nature, and scientists needing accurate data.
One man is a surveyor - so his GPS is as precise and accurate as he says.
From your homepage, click on the Edit Observations tab.
Scroll to the bottom. In the lower left corner, make sure it’s set to “Per page: 200”.
Scroll back to the top. Click on “Batch edit”.
Click “Select All”, then “Edit Selected”.
Click “Batch Operations”.
On the right side of the highlighted area, find where it says “Acc (m):” in small text. Click that. In the small box below, type in your number. I’d use 15 or 20 to be on the safe side, but you can use whatever you’re sure about.
Click the “Apply” button below the location section, then scroll down to the bottom of the page and click “Save All”.
Scroll down again and select Page 2.
Repeat Steps 4 through 8 as necessary. After Page 2, you can just edit the page number in the URL and reload rather than scrolling all the way down again.
I did actually know that, but thanks anyway for the detailed explanation which will certainly be of use to someone . I think though at 200 at a time it would take just too long to go through all my old observations, time which I feel would be much better spent posting new obs or doing some IDing for others. I will, though, try to do better in future!
I used @dianastuder’s link to check my own observations and found 2,156 out of 26,442 have no accuracy recorded. That doesn’t surprise me, honestly, because I often upload between 20 and 60 observations at a time (using the web interface) and put most of my attention on the choice and order of photos, the ID, and the point location; I don’t always take the time to check every parameter. I usually get locations from my iPhone or my Olympus Tough camera, and occasionally they don’t record an accuracy. If they didn’t record one, I would have to make one up, usually well after the fact—and how is that good science? Even when my phone or camera do record an accuracy, I don’t trust it to be…well…accurate, since it can vary widely from photo to photo on the same hike. Lately my iPhone often gives an location accuracy of 1 meter, but that sounds too precise to be trustworthy :)
I’m still rather confused by this obsession with extreme precision in a location. Maybe for a botanist it’s useful to know within a meter or two where that shrub is standing. But for anyone who photos animals, that level of precision is unimportant.
There are millions of specimen records in museums that provide location information that predates iNaturalist. Very few have accuracy (precision) circles associated with their coordinates. Many don’t even have coordinates unless they were assigned later based on location description.
It depends on what your research questions are with the data. With data to within a metre or two, many interesting questions, for example, about habitat use, become possible. Museum data are of no use for such questions, but iNaturalist data have the potential to be useful, if they are collected with precise locations. One day, perhaps, we will look back at the inconsistent and incoherent ways in which we record precision today in the same way that we look back on museum labels that say only “Australia” as a location. All that said, if paying attention to such issues is not your thing, feel free to ignore it. The ideal situation is that very precise locations are recorded by default by the technology and the user should not have to think about it.
My innate sense of guilt (I’m joking, but actually not that much ) got the better of me yesterday and I actually sat down and added location accuracy to all my observations. I feel MUCH better now .
As adding accuracy one-by-one to observations is pretty time-consuming, I’ll try and remember to do a batch edit on new obs regularly. But (just a thought) wouldn’t it be great if those users who habitually use GPS systems (on camera or with smartphone/GPS unit tracking) could set a reasonable default accuracy (say 10m?) for all observations they upload (this could obviously be changed if necessary for individual observations). Or at least if the accuracy could be set for a batch of observations as they’re uploaded?
?) and, like 
. I’ll try and do better in future, but I imagine there is no way of going back and bulk updating more than 7,500 observations, so as a workround, I’ll be adding a line to my profile explaining the process and specifying the accuracy. Not perfect, but I can’t see any other way.
. I think though at 200 at a time it would take just too long to go through all my old observations, time which I feel would be much better spent posting new obs or doing some IDing for others. I will, though, try to do better in future!
) got the better of me yesterday and I actually sat down and added location accuracy to all my observations. I feel MUCH better now 
.