I have multiple point pattern distributions over the same region/polygon. Each point pattern distribution has different number of points.

Here is an example of two spatial point patterns for the region/polygon, the first one covers the polygon, while in the second not all of the polygon area is covered by the point pattern.

enter image description here

How can I compare the distributions within a polygon?

I prefer a solution for this problem in R.

  • What do you mean by 'compare'? There are a lot of ways of looking at point distributions - nearest-neighbour, point densities, etc. – Simbamangu May 14 '14 at 15:34

I would divide the entropy (there is a R package) of the point positions by the entropy of the same number of evenly spaced points inside the polygon. I will not always be possible to have exactly the same number of points, but you can approximate the required spacing based on the area of the polygon (sqrt(num_point/area_polygon)).

Another index could be using a tessalation with your points to create a partition of the polygon, then look at the coefficient of variation of the areas.

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.