I have a feature class that contain many overlapping polygons. Each polygon has a field value which I would like to add together wherever the polygons overlap. There are about 100 polygons each with complex boundaries and some are multi-part polygons. Below is conceptual diagram of what I would like to do.

enter image description here

I know I can do this by converting each polygon to raster and adding them. The problem is that since their boundaries are complex, it takes a very long time to convert. I was wondering if there is a solution without converting to raster?

2 Answers 2


Intersect will give you polygons representing the overlap.

Then you can use Dissolve with a statistic of sum on your value field.

Finally, you can Merge/Append back to the original dataset:

enter image description here

  • So I would need to separate each polygon to its own feature class? Jul 13, 2015 at 19:55
  • @OliverOliver, no you don't need to do it. You can input one or many FCs to Intersect.
    – Paul
    Jul 13, 2015 at 20:05
  • Nice! I did not know this. I'll your method a go and get back to you. Jul 13, 2015 at 20:09
  • OK nice. I think this gets me halfway there. So I dissolved like you said and this does add up the total for each feature within the feature class. The problem there are many overlapping polygons within the same geography ... sometimes there are 10 or more that overlap the same area. So when I dissolve, they go back to the same appearance as the original dataset -> there are hidden geographies because they overlap. Jul 13, 2015 at 20:42
  • 1
    @OliverOliver, you might need to do to a dissolve by field then to separate them. Without looking at your data, it's hard to say.
    – Paul
    Jul 13, 2015 at 20:50

Ok, here is the method I came up with and works for me.

  1. Union feature class by itself
  2. Use Find Identical on the unioned file
  3. Join the unioned file to the table produced in step 2. by the ObjectID and IN_FID fields.
  4. Dissolve by "FEAT_SEQ" field, putting the sum of the score in the statistics field.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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