Accuracy circle can be dragged separately from pin when editing an observation's location

Platform: website

Browser: chrome


Screenshots of what you are seeing (instructions for taking a screenshot on computers and mobile devices:

Description of problem:

In the above observation, I wanted to make the location a little more accurate, which means moving it a few meters. I go to the observation and push edit. For the location, I zoom in and move the location via the red circle to the correct place by dragging the center of the circle, and adjust the accuracy by dragging a dot at the edge of the circle. Everything looks good, so I push “Save observation.” After that, the location ends up being put about 5 km away from the correct location. I redo everything and move the location back to the correct location and push “Save observation.” It again puts it about 5 km from the correct location at seemingly the same exact very bad location where it put it after the first edit.

I’m trying once again to edit by dragging the red circle, but notice the lat/longs don’t change as I drag the center. The accuracy changes though as I drag the edge of the circle.

Can you share some screenshots, e.g. of the observation edit screen, including where you moved the pin to vs. where it ends up?

I tried editing the location manually of an observation and the lat/long do adjust when I click a new location on the map or drag the circle to a new location. The changes save as expected as well, so some more, excruciatingly detailed information might be necessary to help troubleshoot.

1 Like

I was able to correct the location by going to google maps to get the lat/longs and manually type them in. I’ll try to duplicate the problem with screen shots

Here is the correct location: image
Starting to edit:

New location as a test:
After I push “Save observation”: image
This location is not horribly wrong, but it is still slightly wrong. It’s not where I put the center.

1 Like

I also notice the red dot in the right part of the green circle is the same observation as that of the green thingy in the center of the green circle (which is the current problematic observation). In other words, the observation shows up as two locations on the map.

The orange pins on the maps are cached, so you can ignore that (that is, it’s temporary: the orange pin will eventually move “underneath” the green one in a few minutes).

I’m able to replicate the issue if I click and drag the white, circle center point. Usually I grab the red pin and not the white circle, so I’ve never noticed this before. Essentially: grab the red pin instead of the white circle and you should be able to update your locations correctly. But I would consider this a bug since it shouldn’t be possible to move one and not the other.

It seems to be a slightly different format for selecting locations on the main website upload page (no red pin), so when the Edit observation page is updated, perhaps that will be standardized?

web uploader with no red pin:


that’s interesting. i was trying to reproduce this problem the other week (, and i couldn’t.

i guess the difference between what you guys are doing and what i was doing is that you’re clicking on the edit link first. i tend to ignore clicking edit first. so i’m not even presented with the option of clicking on the white circle. but making a coordinate change by dragging the pin the without clicking edit first seems to wipe out the positional accuracy. so another workflow variation to deal with, perhaps, if this screen is to be fixed?

(I don’t usually click “Edit” either. But the white circle will appear, after you click the red pin, even if you don’t first click Edit.)

The web uploader maintains the accuracy when you move the center around around (to a hilarious number of decimals). A single “center” also seems better. I dunno if that “undo” button is really necessary tho I’m sure someone uses it.


I never noticed that it the Location Accuracy went up to fifteen decimals! Now that is what I call sub-atomic precision! Is it really necessary for the Location Accuracy to have that many decimals?

Made an issue here and changed the topic’s title to be more descriptive.

We’re big Ant-Man fans here at iNat. ;-)


So its pym particle precision?