I am trying to generate a point feature class of possible plot centers for a research project. The plot centers will be inside of defined forest stand polygons which are aggregated in various ways across the landscape. What I am trying to achieve is this: approximately (+/- 100 m) equidistant points within and between forest polygons, so each polygon has two points inside and then one or both points are the same distance from a point in a separate polygon so that each pair looks something like this (the green/orange shapes are supposed to be two forest polygons, the blue points are the plot centers, and the red lines are just to show equal distance between points-I don't need to actually create a line class connecting the points, the euclidean distance between them just needs to be similar.

enter image description here

I would also like to be able to define a minimum buffer distance between the points and the polygon edges. I have lots and lots of polygons, but for most of them this configuration will not be possible, so I'm looking for a way to identify potentially usable polygon pairs. I have access to ArcGIS Pro and I use ArcMap over Citrix if that's relevant. I have limited experience with Python but am not afraid to learn if that's the best solution.

  • 2
    Ouch Offhand, this looks like an NP-Hard problem. An automated means of choosing two points inside a polygon approaches NP-Hard all on it's own; two such sets, a fixed difference apart, cannot be easier. Identifying polygon pairs a minimum distance apart shouldn't be all that difficult, albeit time consuming and possibly exponentially complex. Start with the Near command (if you have an Advanced license), otherwise you're about to get saturated in Python coding.
    – Vince
    Dec 28 '18 at 20:03
  • Welcome to GIS SE! We're a little different from other sites; this isn't a discussion forum but a Q&A site. Please check out our short tour to learn about our focussed Q&A format.
    – PolyGeo
    Dec 28 '18 at 22:47
  • 1
    Unfortunately, this problem isn't a good fit for our "Focused question/Best answer" model. There are tons of issues to discuss, but this isn't a discussion forum. While we do have Geographic Information Systems Chat, it is underutilized, so a different environment would be needed.
    – Vince
    Dec 28 '18 at 23:09

I get from that:

enter image description here

to that:

enter image description here

with no scripting, although results can be improved with some. The key here is exponential reduction of chances to find solution with changes in distance between points. I expect you'll find very few suitable pairs.

Let me know if this is what you thinking about and I'll post workflow.

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.