Polar Observations (and the limits of the system to display them)

General
1

Earlier today I happened upon this excellent observation of a South Polar Skua at -81.66 Latitude. It has 115 faves and 150 IDs because in addition to being inherently fabulous, it is by far the southern-most observation on iNat’s Explore map. https://www.inaturalist.org/observations/133388276

This made me wonder about northern-most observations, and using the explore map, this one: https://www.inaturalist.org/observations/299568624 seems to be the northern-most, at 85.2 North. But looking at that one on the map, I noticed two things. First, that the base map cuts off at 85 north and no observations are displayed beyond that line. Second, that the Polar Bear observation listed above displays only when zoomed out enough to be part of an area, not as a single observation marker, and only below the 85N line.

This made me wonder if there were more observations north of 85 that just weren’t displaying. So I edited the URL coordinates to get this search: https://www.inaturalist.org/observations?nelat=89.99999999&nelng=180&swlat=85&swlng=-179.99 and found out there are a lot of observations up there, currently 111. Most of these are clearly incorrect locations, but there are a lot of honestly North of 85 observations of seals, bears, and birds. That Black-legged Kittiwake really could be at 88.47N. And the Hooded Seal is likely really at 88.8N. But this polar bear https://www.inaturalist.org/observations/181327781 at 89.989 is likely the northiest observation so far.

This, of course, led me to look into whether there are a similar set of observations closer to the South Pole than -81. And it turns out, there are several. Many of these, again, are incorrect locations (“off the Antarctic Peninsula” is not -82.86 Latitude. But there is also one that is honestly just too far south (-84.4) to display on this map: https://www.inaturalist.org/observations/78895727

So the map was just hiding the northern-most and southern-most observations, along with a bunch of observations that have incorrect IDs, in some cases by several thousand miles.

2

I suggest that observations farther north or south than some limit (85° or 80°, whichever) be displayed on a stereographic projection instead of Mercator.

3

It would be cool to see a system that could transition between projections depending on the geographic focus.

4

GBIF does a couple of things that are interesting:

  1. it plots the observation on a 3D globe view:
    image
    image388×383 22.6 KB
  2. it uses a different projection for 2D maps that seems to show the poles better than web Mercator:
    image
    image745×730 118 KB

the second thing would probably be a bigger change for iNaturalist, but i think it could be relatively straightforward to add a globe view.

here are some iNat observations near the north pole, with GBIF links: https://jumear.github.io/stirfry/iNatAPIv2_observations?quality_grade=research&nelat=90&nelng=180&subview=map&swlat=86&swlng=-180&options=outlinks&outlink_source=GBIF

5

then we would like an equal area projection :grin:

https://forum.inaturalist.org/t/world-map-representation-is-mercator-the-most-adequate-for-inat/7806/3

6

I like equal area projections! It does make me dream of much less sensible projections. Like one where every landmass is scaled according to the quality of the pizza. Or an equal volume projection, where landmasses are scaled based on their area times their average elevation.