We have a traditional project limited by the boundaries of Florida (place_id=21). The problem is that we are not getting observations for corals and other taxa that are just outside this boundary, with similar at the border of adjacent states when the 30km taxon geoprivacy obscuring comes into play. I tried creating a generalized polygon with about a 50m buffer to the north and well out into the ocean around the coastline and ran into this issue https://forum.inaturalist.org/t/inat-times-out-when-trying-to-load-place-polygon-504-gateway-time-out/35291.
So what I’m wondering is whether anyone has a solution, such as maybe an already existing place that would work or whether there might be a way to petition iNaturalist to create a place or anything else that I’m not thinking of. Any help would be most appreciated.
You can actually create a place yourself, with a KML file (you can make one for free on Google Earth). I would not recommend making another place for Florida, as it would be confusing, but maybe one could be made for “waters around Florida” or something. A project could then include the locations “Florida” and “waters around Florida.”
I’d suggest removing the “must be observed in Florida, US” rule from the project - then people could add observations from the oceans off of Florida. You might get some stray non-Florida observations added but those could be easily removed, and I think it’s obvious the project is just for Florida and not Georgia, and elsewhere.
As long as you keep this part of the project intact, the boundary will show on the map to help people understand what the project is for, but observations won’t be restricted to that boundary.
Is there any way to petition iNaturalist to create a place such as I’ve described if users are not allowed to? And, related, it would be really nice if the documentation about creating places was changed, as I kept trying and trying, assuming that iNat servers were just overloaded or something, since ther was nothing saying that there was a limit on how many records a polygon might encompass.
I find it easier to create the boundaries in Google Maps, then export it to KML for import to iNaturalist.
iNaturalist Places have a 1MB limit for the KML import. If you’re creating an extended area directly off the official boundaries of the Florida land mass, you’ll have a complex, and large, KML file. It’s also possible that it falls within the 1MB limit, but is too complex for iNat to ingest. In either case, iNat will choke on that with the 504 error you encountered.
In such a case, I create a simplified boundary I draw by hand in Google Maps using the polygon tool. You can customize that boundary in any way that makes sense, for example: keeping tight to the borders of adjacent states, but extending out as far as is needed to encompass coastal areas of interest. Where iNat Places already exist - e.g.: Dry Tortugas National Park, Biscayne National Park - you can just add those to your Project.
Neither the file size (1.79kb) nor complexity of the poly is the issue. The issue is, as in the link in the original post, that there are too many observations within it. I could not find any way to add more than one place, but if you know how, please let me know.
Not really. Standard places won’t get changed until/unless we change them all, and I doubt if/when that happens they’ll have more than the current coastal buffer. What you can do is add some small places that capture the areas you want and include those in the project.
You have to have multiple “Must be in place” rules and you can only add one rule per save. So add a rule, then save, then edit, then add another rule, etc.
So I tried creating a polygon for the ocean around Florida, with the thought that there wouldn’t be many observations captured, but it still timed out. I then tried to do the same for just ocean off the NW of FL and that still didn’t work. Neither poly was anywhere near the size limit and I can’t imagine that the NW offshore one especially captured many observations, so can you tell me what the limits are and/or why these are failing when I try to turn them into places?
The solution to this appears to be to to create smaller and smaller multiple barely-overlapping polygons until iNaturalist would accept them and then adding them all to the project, although I haven’t finished doing this yet, in part because I can apparently only create three per day.
