I have polygons for the country divided into E.D. areas.

These polygons illustrate the average annual rainfall in their ED area.

Is there any way to statistically tell if one polygon area has a statistically higher or lower volume of rain compared to its surrounding areas?

or if I extract points from these polygons and make a smooth surface raster via interpolation - is there a way I can I assess my layer for statistically higher or lower rainfall ?


I have a data set of polygons. Each polygon is an area in the county and each polygon has an "Average annual rainfall in mm.".

The range of values is between 1-20 mm

For example, If I had a polygons or even a raster grid as follows:


In the second line the number 12 is much bigger than any of the surrounding values or any of its neighbours. However, the values range from 01-20 so it is not large by comparison to the rest of the dataset.

  • 1
    $\begingroup$ Could you tell us what you mean by "statistically higher or lower"? After all, presumably the average rainfall is whatever it is and all you have to do is compare one number to another to see which is greater. And do you really mean "volume" or just the usual depth, which is volume per unit area? What do you mean by "extract points" when you have described your data as given as an average per polygon? $\endgroup$ – whuber Sep 9 '18 at 18:18
  • $\begingroup$ I'm afraid you have answered my question about your meaning of "statistical" in a circular fashion, by stating it means "in a statistical sense"! Could you supply an explanation that adds some information about your intention? $\endgroup$ – whuber Sep 9 '18 at 19:04
  • $\begingroup$ @whuber aplogies - that was not the complete reply. I ran out of characters and have amended the question instead. The rain data I have is annual rainfall, corresponding to the volume of rain each district area receives annually. By extracting points, I mean extracting the centroids of the districts and using these points to build a raster using interpolation. $\endgroup$ – Will.S89 Sep 9 '18 at 19:07
  • $\begingroup$ @whuber - update: I think I may have found the solution : The Anselin Moran's I. $\endgroup$ – Will.S89 Sep 10 '18 at 20:18

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.