2

I would like to perform a spatial join using a (non-existent) "contains part of" match option. i.e. I want the join to include any and all polygons that overlap, not just touch at the edge. None of the match options available produces satisfactory results.

To clarify, in the example below, the yellow polygon should have a join count of 4 (A, B, C, D).

Spatial_join_example

INTERSECT gives a join count of 6 (A - F)

The result I am interested in is the the join count per polygon

Any ideas for a work around or is there another tool that will do this?

  • 1
    I'm not understanding, why doesn't the intersect operation work for you? – Dowlers Nov 19 '15 at 18:18
  • 1
    Please edit your question to more clearly define "contains part of" and "join count" with a graphic example embedded in the question as an image. – Vince Nov 19 '15 at 18:34
  • like @Dowlers says - I also think the intersect tool will give you the results you need. – MDHald Nov 19 '15 at 20:46
  • I have edited the post to clarify why INTERSECT doesn't work for me – Tim Nov 20 '15 at 11:27
  • Thank you all, using the "intersect" tool worked rather than using spatial join with the intersect match option. Join counts were calculated from unique ID frequency. – Tim Nov 20 '15 at 15:05
1

As mentioned by @Dowlers and others, intersect is a key here

  1. Intersect itself (1 layer only!)
  2. Calculate frequency of polygon uniqueID on output from above

Result:

enter image description here

Not applicable for big polygons which completely contain others

  • Can you clarify what you mean by "Intersect itself (1 layer only!)"? when I use intersect the lowest polygon in your example (AA01, brown outline) would have a join count of 2. – Tim Nov 20 '15 at 11:29
  • Thank you, I realised the "intersect" tool is what I needed to use rather than a spatial join with INTERSECT match option. As you suggested I used intersect on the two layers, removed duplicate rows and counted the frequency of FIDs from each layer to get join counts. – Tim Nov 20 '15 at 14:59

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.