1

I have a point shapefile "Negative_vel". With the command Analysis tools-Proximity-Buffer and selecting 50m as a linear unit and dissolve type ALL I created a new .shp "Negative_vel_buffer50m". But this new shapefile (with the buffer circular polygons) has no attribute table. I need an FID for each circular buffer polygon so that to check with Join Points to Layer how many points from the initial shapefile "Negative_vel" are within each buffer polygon. Because I need to keep only the polygon buffers with more than 3 points inside them. It is not of my interest to keep buffer polygons with just one point inside.

So, I need a buffer .shp dissolved with attribute table so later on to Join and count the points inside each polygon.

  • 2
    You should specify which software you are using. The first thing that stands out is you buffered with a dissolve type of all, meaning that it creates a single, multipart feature consisting of all buffers and eliminating any borders where they overlap. Are you sure there's no attribute table? Or does it just have one record? You want the dissolve type set to NONE when you run the buffer if you want distinct features (and hence FIDs) for each buffer. – Chris W Feb 11 '15 at 21:05
  • I'm using ArcGIS Desktop 10.0 – Chris Feb 11 '15 at 21:43
  • If I select DISSOLVE NONE then I get one buffer polygon per point whereas I want for close points (distance<100m) to get them in one buffer polygon. In my case with Dissolve ALL there is an attribute table but just with one record. – Chris Feb 11 '15 at 21:46
  • 2
    That's something different than stated in your original question (you should edit your question with the button at the bottom left of it to add clarifications) and requires an extra step or perhaps an entirely new solution. You may be able to take the buffer you created and Explode or Multipart to Singlepart it. The result is any buffers that don't overlap become their own shapes. Since your buffer is 50m, any points within 100m of any other point should be in the same buffer, but there could be points in that buffer farther than 100m apart. It sounds like you're trying to do clustering. – Chris W Feb 11 '15 at 22:21
3

In the Buffer tool there is no Just Overlapping option, it's all or none. To make the buffers merge where overlapping and nowhere else it's necessary to buffer with dissolve of 'all' and explode the feature using Multipart to Singlepart.

Now each cluster of buffers is a single polygon and you can use the spatial join tool to count the number of points in each polygon by transferring the polygon FID/OID to the points and use Summary Statistics to summarize the records (min/max/count of each Case polygon FID).

Hint: to make it easier to understand what's happening in the joined table I suggest to add a field to the points (PointID/LONG) and calculate the FID/OID, do the same for the polygons (BufferID) before the spatial join so that when the table is joined you will have a field PointID and BufferID to clearly recognize which ID came from where.

1

The Spatial Join tool will allow you to count how many points are within each of the buffers.

Do you also need to have the original ID's of the source points attached to the buffer polygons?

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.