I am trying to work with some data from Antarctica in QGIS. I have defined this new projection:

+proj=laea +lat_0=-90 +lon_0=0 +datum=WGS84 +units=m +no_defs

And started adding a few vector layers. One of these contains the location of meteorologic stations in the continent. No matter what I try, QGis never portraits the South Pole station:

enter image description here

Is this the result of an issue with the custom CRS or is it a bug with QGis?

  • 2
    It's not a problem with your CRS, I get it too - it seems generally flaky overall - often other points don't show either - but interesting that you don't see that? You'll probably find it works if you reproject your points layer so that it's not "on the fly". (Are you using "on the fly"? Also, what is your point source - I used CSV). I tried a few points directly off centre, at different longitudes, and slightly north of SP too but no better. I get a weird "level of detail" glitch when zoomed out too, but that's going to be dependent on the data layer as well as the graphics hardware in use.
    – mdsumner
    Commented Apr 8, 2015 at 10:16
  • As you can see from the image, the stations layer was imported as a CSV file containing latitude and longitude. I shall try to reproject these points into a new layer, if it functions then this is definitely a bug. Commented Apr 8, 2015 at 11:44

1 Answer 1


As written in the comments, it is better to reproject your data into exotic target CRS and don't rely on the OTF reprojection of the software:

enter image description here

For raster data, you may append -wo SOURCE_EXTRA=1000 to the gdalwarp command line to avoid nasty artefacts along the 180° meridian.

  • Didn't know that gdalwarp trick. Commented Jan 3, 2016 at 8:49
  • See my answer on gis.stackexchange.com/questions/173766/… and gis.stackexchange.com/questions/126747/…
    – AndreJ
    Commented Jan 3, 2016 at 10:13
  • Could you explain what is an "exotic target CRS"? Commented Jan 3, 2016 at 17:44
  • Because all meridians touch in one point. You would not see that in "normal" mercator or transverse mercator projections of populated parts of the world. There you only have more ore less bended lines. Mathematically, this is far more simple than mapping the poles correctly.
    – AndreJ
    Commented Jan 3, 2016 at 18:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.