the equator is ~40000km, so if you have a circle with radius ~7000km, you’re going to have a circle that covers almost an entire hemisphere. i bet you run into issues trying to visualize a giant circle like this when it encompasses one of the poles.
in cases where you have a radius of 39000km, that doesn’t even make sense. so in those cases, i bet what happens is the mapping platform doesn’t even try to visualize this and just defaults to the whole world.
I guess it could be a choice to not show confusing shapes like those, but I think it’s still useful to be able to see which hemisphere it’s in if there aren’t helpful locality notes.
when the map tries to jump to a particular spot, it will try to zoom to the proper extent, it looks like that just isn’t handled well when you’re encompassing the poles. (the web mercator projection isn’t really designed to display areas within about 5 degrees latitude of the poles.)
i pinned location with a giant accuracy radius, and then i selected the pinned location. here’s how the map on the upload page tries to visualize that extent:
notice how the center of the viewport is far from any area covered by the basemap. most of it appears just blank. that’s effectively what you’re seeing in your observations.
so if you open your humpback observation, you’ll notice that you can drag the map down a few times, and eventually you’ll see a recognizable basemap:
so again, i don’t think there’s really a bug here. it’s just the weirdness inherent in data like this when it encompasses the poles on a projection that isn’t designed to handle polar locations.
