Altitude / elevation for coordinates for alpine species

[rukayaj]

I'm at the Uzbek CESP workshop and we have some participants asking if there is any possibility for GBIF to add a flag for alpine species when the provided altitude is incorrect (based on the coordinates). I was explaining how to do this through QGIS or a GIS program, but I think it would actually make a lot of sense if there was a flag on GBIF for these kind of suspicious records. Is something like this feasible?

[Mesibov]

Please see https://discourse.gbif.org/t/a-guide-to-elevation-issues/4375. A flag based on elevation above local mean sea level would mean that millions of GBIF records would be flagged because the data providers used elevation above the GPS'es ellipsoid. Another problem with alpine records is that in steep country a small coordinate uncertainty (horizontal) can mean an even larger elevation uncertainty (vertical).

[Mesibov]

@rukayaj, I recently checked a large dataset with a full complement of verbatimElevation entries. The check wasn't to compare vE values with correct ones for the coordinates, but specifically to find repeatedly sampled locations where the difference between the lowest vE value and the highest was greater than 100 meters. (The code I used is explained in an upcoming BASHing data 2 blog post for 24 May.) There were a lot of finds, unfortunately. I attach a screenshot of some of them. Field 1 is the range in meters, field 2 is the record ID (a UUIDv4), fields 3 and 4 are the original latitude and longitude, field 5 is the elevation and fields 6 and 7 are the lat/lons rounded to 4 decimal places.