I would like to change a distance unit on Euclidean distance analysis. How can i change the current unit to feet or meter?enter image description here

what unit is the unit for 0 and 0.18939. is it degree?


Yes, it's degrees, because your raster is in geographic coordinates. Project the input data, either the raster or the features that made it, to a suitable projected coordinate system (units feet or metres) then rerun your Euclidean Distance and your units should be in the same units as the features.

The environment output coordinate system has an effect on the created raster and should force the results in the units of this environment if set, but either way by projecting or setting the environment it would be best to recreate the Euclidean Distance rather than trying to project the existing raster as resampling will warp the distances and they will no longer be reliable.

  • Not OP but thanks. What shall be a good meter-based coordinate system to use for the output of Euclidean distance? – AndrewLebron Nov 5 '18 at 18:26
  • It depends on where your data is and how big the coverage area @AndrewLebron. If you're only a few (hundred) kilometres wide or less then UTM should work well if you're not crossing zones; you can go slightly outside the zone but the distortion increases the further your data extends beyond the 6 degree zone. If you have very large data then a Lamberts conic or Albers equal area would be more suitable. – Michael Stimson Nov 5 '18 at 23:12
  • It's just one country: Nigeria. I'm using WGS 1984 32N. Seems to be coping well. – AndrewLebron Nov 5 '18 at 23:51
  • That sounds like a good choice @AndrewLebron, see epsg.io/32632 WGS84 UTM 32N (EPSG:32632) is a suggested coordinate system for Nigeria but doesn't cover the whole country, Benin City is on the boundary of the zone in the west and Numan or Biu on the east. As I said though you can go slightly outside the zone boundary without too much distortion just beware the far west of Nigeria would be better in WGS84 UTM 31N (EPSG:32631). – Michael Stimson Nov 6 '18 at 0:06
  • Thanks a lot for the additional information and help. I mostly want the result of my raster outputs to be measurable. I'm trying to pick an optimal site. And one of the steps is to reclassify some euclidean distance results (less than 2,000 m is a score of 10 etc). And degrees output doesn't do me any good in that case. Now I feel way better. Thanks. – AndrewLebron Nov 6 '18 at 0:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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