1

I have created a buffer of 150m around a point. This should give (a polygonal approximation of) a circle with radius 150m.

I tried to verify this with the measurement tool, and I am baffled to see a difference of ~30% in length, depending on whether I choose the "Cartesian" or the "ellipsoidal" measurement. The first one gives the expected 150m, the second one only gives 103m.

I think:

  • The difference between Cartesian and ellipsoidal measurements can never be this big on such a short distance.

  • Ellipsoidal distance should always be greater than Cartesian distance.

My layer CRS is "EPSG:3857 - WGS 84 / Pseudo-Mercator".

Am I thinking wrong, or is this some error in QGIS?

Screenhsots of the measurements (sorry - only in German):

cartesian measurement

ellipsoidal measuremnet

Vince
  • 20,017
  • 15
  • 45
  • 64
AEF
  • 123
  • 3
  • 6
    Never ever try to measure distances and areas using Web Mercator, because of the effect you're observing. Choose a different, local, m-based CRS and try again. – Erik Oct 04 '23 at 13:25
  • 2
    OK, good to know. Do you know the reason why? Which of the measurements is correct? – AEF Oct 04 '23 at 13:27
  • 1
    Web Mercator is *never* correct on distance. – Vince Oct 04 '23 at 14:53
  • 1
    Consider an equal area projection for your analysis. https://gis.stackexchange.com/questions/209019/what-is-the-ideal-equal-area-projection-to-map-germany-switzerland-and-austria – GBG Oct 04 '23 at 15:06

0 Answers0