I have a dataset of many (>13 million) points, representing borders of potential quick clay zones, and need to enclose them exactly by polygons. Each cluster has an ID, however, the tools "minimum bounding geometry (convex hull)" and "concave hull" yield larger areas than the minimum which would enclose the points. Is there a more precise way to get the minimum bounding area?

(Basically, I want the red areas removed)


  • 2
  • @Taras the problem is, they are (and need to be) grouped by ID, I didn't find a way to group them with alpha, or did I overlook something? May 10, 2022 at 13:30
  • 1
    You are correct, there is not grouping field in the "Concave hull (alpha shapes)" tool. You can either: (1) Collect geometries by "ID" and apply the Concave hull with iterating over each feature or (2) Split features by "ID" then apply the Concave hull and after merge the outputs from the previous step
    – Taras
    May 10, 2022 at 13:46
  • @Taras thanks for the tip! Splitting features isn't really an option for me, since I've got some 10'000 areas, but I'll try collect geometries -> concave hull/alpha May 11, 2022 at 6:01
  • 1
    @BERA same cluster, I tried coloring them different, but that was too much for the PC^^ May 11, 2022 at 6:01


Your Answer

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

Browse other questions tagged or ask your own question.