I am currently dealing with a dataset which consists of about 2700 points stochastically placed in a study area of about 650,000 sqKm. I am trying to create a distance raster for these points, which I have been able to do using Euclidean Distance tool, but I would like this distance raster to also be affected by the class of point being evaluated. For example the points range in value from 6 to 10. Therefore I would like the raster to be affected by these values. So for points with a high value I would like the distance value to be lower as to positively affect a Cost raster and eventual least cost pathway.

I know that this is a little vague, I apologize it is because I do not know what this concept would be called. Any insight would be extremely helpful.

  • If you have 5 classes, could you create 5 euclidean distance rasters on each of the 5 classes, then divide each one by the 'weight' you're describing for each class, then merge them all back together? – DPSSpatial Mar 24 '15 at 20:50
  • Distance is distance, so I'm not sure what you mean by lowering the distance around certain points. I understand you want to create a cost raster based on distance, but with weighting. At that point it's no longer a distance, but a cost value that happens to equal distance in some places. What is the relationship between the two? I mean if you ED the points you get zones around each point (increasing values until they meet the increasing values from the next nearest point), and you want to lower the values of a zone - none for 6, 1 for 7, 2 for 8, etc? Or some other formula? – Chris W Mar 24 '15 at 20:57
  • 1
    @mapBaker I believe this is what I am looking for. It would make sense to separate them out as classes and then weight them and merge back together. – rgmrtn Mar 24 '15 at 22:03
  • It is a little vague. Could you reword/rewrite, via the edit button for clarification? Maybe incorporate some of Chris' and MapB's terminolgy. – Martin F Mar 25 '15 at 14:42

Your Answer

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

Browse other questions tagged or ask your own question.