I have a points shapefile where in any given location there may be one point, two points, or three points on the exact same spot.

Any suggestions on how to count those up so I get one point at each location, but a count of how many points were there before?

  • The definition of "same exact spot" can be somewhat fluid with respect to point/point comparison. The usual methodology involves small buffer circles (within the precision of the dataset -- usually 1-10 meters). Remember to add an ID test to the query, so that only those queries where the minimum ID value is the query feature are actually output. – Vince Mar 3 '17 at 2:42
  • Do you know python? – Pura Vida Mar 3 '17 at 2:42
  • 1
    The shape field (with tolerance) can be used in the Find Identical tool, desktop.arcgis.com/en/arcmap/10.3/tools/data-management-toolbox/…. See particularly the last Python example on that page for 'counting' the duplicates. Then there is the Delete Duplicates tool for similarly targeting those duplicates for deletion, and this operates on the input data. – T. Wayne Whitley Mar 3 '17 at 7:06
  • @Vince true, thanks. In this case, it's actually exact to within machine precision due to how they were created, but I may just do the little buffer strategy. Thanks! – nick_eu Mar 3 '17 at 20:30
  • @T.WayneWhitley thanks, but my sense was "find identical" was for identical field values, not identical locations, no? – nick_eu Mar 3 '17 at 20:31

Any suggestions on how to count those up so I get one point at each location, but a count of how many points were there before?

This can be done using the Dissolve tool on the layer, leaving all the Dissolve Fields unchecked, and adding ObjectID as a Statistics Field with COUNT as the Statistic Type. As has been pointed out though, ArcGIS' definition of the "exact same spot" may differ from yours depending on what you're doing.


After a while I just decided to export the points to R and counted duplicate coordinates (since it's just a point file, it translates into a tabular dataset well). Best in-arc strategy is probably @Vince's suggestion of just little buffers.


I found a solution that might be helpful in these situations.

1 >>>First Use buffer tool to create a small polygon on every point of your layer. CAREFUL! set the Dissolve type on all. or you will have multiple features in the same spot and will make a lot of trouble later! Choose a small radius unless you want to join a small cluster of data. for example, your points are not exactly in the same spot and may differ a bit.I used 0.5 meter radius. even 0.1 meter is enough if the points and exactly the same

2 >>> Use Feature to Polygon Tool to split all the polygons and eliminate all multi-part features.

3 >>> open Spatial Join and put your points as Traget and the buffer as join feature.

4 >>> Now your points in same location have Identical join FID Field. So you can dessolve your points based on this field. open Dissolve (Data Management), your points as input, check Join FID as dissolve field. then on statistical field choose Object ID and select count as Statistic type value.

5 >>> the result hase your every point with number of those have identical locations in Count ObjectID field.

I used it to analyse the intersections and how many roads lead to an intersections.


Could just use the collect events tool?


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.