I am currently working on my thesis and after trying for a few days I decided to ask this questions since I guess I'm stuck. I'm working on ArcGIS PRO 2.7.0 with datasets obtained via Esri (online) the Netherlands.

Basically, I have data from a lot of buildings all depicted by polygons which is one data layer (I have clipped it to the municipality of Amsterdam and hence it contains around 120000 rows). In addition I have outlined a few parks in Amsterdam I wanted to research and I have made buffers of 10~50m around them. What I want to research is the functions of buildings in close adjacency to parks.

I have tried to intersect the buffer layer with the buildings layer which has given me outcomes as expected. However, I want to research only the FIRST ROW of buildings close to parks. Is there a way to do this (non-manually)? I have attached a screenshot of the outcome of the buffer (40m) and buildings layer in which you can clearly see multiple rows of "buildings" are selected. To clarify, I am trying to obtain an output in which only the first row of buildings (i.e. the buildings that are adjacent to the parks) are selected; the buildings that are surrounding the park on each site.

I am thinking that an option could be to change the polygon layer of the parks (indicated in black in the picture below) to a line. I read somewhere that this could help but I am not sure.

enter image description here


2 Answers 2


The problem here that needs solving is due to the fact that the rows of buildings are different distances away from the park boundary. For example the buildings to the south have a road in-between them and the park. So the distances are not constant. Also the number of buildings per side varies. Finally the configuration of buildings varies. In the north you have linear stacks going away from the park boundary and even changing in size, whilst other side's run parallel. You also have different adjacency rules, the west and south sides have buildings touching each other whilst the buildings in the east are detached.

Running the near table tool will also yield incorrect values as a second row of buildings is still closer than the first row of buildings which are further away.

You say you have only a few parks, the logic to process this when everything is an exception to everything would be very difficult to codify if not impossible. I recommend you do this manually.

  • Hi Hornbydd thanks for your reply! I haven't tried the table tool yet, but I won't now if you say it is no use. I have around 20 Parks in total (but I want to scale this up to other cities in the NL) so I guess manually it is doable but not preferable. However, let's say I were to do it manually, you recommend me just selecting all the different buildings adjacent to the parks and make a new data layer for it?
    – Floris508
    Commented Dec 24, 2020 at 12:06
  • 1
    As a software engineer, I have to say that very little is impossible, but if you can't generate a manual heuristic, you aren't ready to start automation.
    – Vince
    Commented Dec 24, 2020 at 14:02
  • Entirely depends upon how you manage your data, an alternative that keeps your building data altogether is to add a new field and tag, a simple select and update.
    – Hornbydd
    Commented Dec 24, 2020 at 14:14

In the end I ended up using the following tools 1) Densify, 2) Feature Vertices to Points and 3) Near tool. Although it is not perfect it is a more precise way than buffer analysis.

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.