1

I have two separate datasets ('Wards' and 'Census Output Areas'), whose boundaries don't quite match up (see below - wards in red, COAs in grey).

BoundaryErrorImage

Is there a way of automatically adjusting the Ward boundaries so that they align with the Census Output Areas? I know that I could use the 'Align Edge Tool' on the Topology toolbar to fix individual errors, but it would take a long time to do them all. After reading about topology rules (here), I think the most appropriate rule is 'area boundary must be covered by boundary of', but I'm not quite sure how I would use it and it seems there are no fixes for this anyway.. Any suggestions?

4

Topology tools may not be necessary in your case. I would transfer the Wards attribute (feature to points, then spatial join) to the census polygon, then dissolve based on the Wards unique identifier. This will yield a new feature class with the attributes of Ward and the boundaries of the Census.

  • Brilliant! I hadn't thought of that - thanks very much – user32476 Jan 8 '16 at 11:35
  • If I'm not mistaken, that suggestion has flipped the wards and COAs. Since the COAs are nested within the wards, you would need to convert the COAs to points; otherwise, the ward point would only join to one of its nested COAs. Maybe I'm missing something. – Tom Jan 8 '16 at 15:47
  • exactly: convert coa to point, spatial join points with wards, table join coa with the points attribute, dissolve. – radouxju Jan 8 '16 at 20:07
0

Take a look at this tutorial: http://webhelp.esri.com/arcgisdesktop/9.3/tutorials/editor/Editor_Ex7_1.htm

It explains how to create a topology and set topology rules. Be aware that both features needs to be stored in a GeoDatabase.

  • The tutorial is useful, but I'm still not sure how to use it for this particular case. I've resolved it now by doing a spatial join instead! – user32476 Jan 8 '16 at 11:37

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.