Several times I have seen observations where the accuracy is rather low, ca 2000km. Since I am tracing certain taxa in certain localities, I am pretty sure the accuracy was much more reasonable at the moment of uploading. The accuracy dropped later. I do not think it was done by observers, and suspect a bug. Did anyone experience that issue? Thanks.
Please fill out the bug report templateandr provide examples - otherwise it won’t be possible to address something like this…
A recent example was https://www.inaturalist.org/observations/167950457 – I wrote “was” since the observer had already corrected the accuracy. AFAIK the accuracy was fine at the moment of uploading, and then changed to 2000+ km.
I do not think I am able to find other examples right now but I have experienced the same several times before.
I did not write about the platform etc. since it seems that it’s not a platform issue – it’s rather a database issue. I.e, some spontaneous changes happen. I wrote this hoping that other users will write about they experience, so that the picture become clearer.
Ok, I’ve switched this to the “General” area as it isn’t currently actionable as a bug report. If more details come to light, it is possible to make a bug report using them.
I think it’s the same as this bug? https://forum.inaturalist.org/t/location-accuracy-expands-when-opening-saving-location/16225
I think no. There, it is about x2, and in my case(s) it is about x10000. :/
platform does matter because different platforms provide different ways to input the accuracy value. in this case, it looks like the observation was originally created in Android. in the Android app, you’re probably most likely going to get a large accuracy value as a result of not zooming in very far on the map when you manually select your location. the bug that spiphany noted could also cause large accuracy values even when the user does zoom in.
I use website version only. Thus I have no idea what is seen at other platforms. I suspect the same.