Check if polygons overlapping another by same attribute

625
3
10-24-2012 08:04 PM
AndreasDaim
Occasional Contributor
hey,

is there a possibility to check, if polygons from one feature class with the same capture date (attribut) overlap each other?

attachment:
captured influence polygons (blue, black..) on different days with GPS Mobile Handheld and therefore got once uppon another laying polygons which must not overlap each other in case, they are captured on the same day = attribute date.
[ATTACH=CONFIG]18701[/ATTACH]

king regards
andreas
0 Kudos
3 Replies
T__WayneWhitley
Frequent Contributor
I'd say the easiest way is to run a Frequency (or Summary Statistics) on the capture date (depending on how it is formatted), export the duplicates and set up a topology enforcing the rule 'Must Not Overlap'.  Of course, you may have to consider tolerances on neighboring polys?...If that's the case, you may try something else; if overlapping polys are generally the same shape/area, maybe Feature to Point (again, for duplicates only), then analyze how close together points are.  Those in close proximity are suspect and warrant closer inspection.
0 Kudos
by Anonymous User
Not applicable
Original User: w4ranger

I'd say the easiest way is to run a Frequency (or Summary Statistics) on the capture date (depending on how it is formatted), export the duplicates and set up a topology enforcing the rule 'Must Not Overlap'.  Of course, you may have to consider tolerances on neighboring polys?...If that's the case, you may try something else; if overlapping polys are generally the same shape/area, maybe Feature to Point (again, for duplicates only), then analyze how close together points are.  Those in close proximity are suspect and warrant closer inspection.


hm, don't really get the idea with the Frequency?

my set up is like this: there are polygons caputerd every few days over a period of 8 months with the exact date.
So I would have hundreds of layers if I would seperate them by date. And on one location there might be polygons overlapping every few days. And they just must not overlap within a capture event. Tolerance has not to be considered.

gz
andreas
0 Kudos
T__WayneWhitley
Frequent Contributor
So you want to identify which (if any) polygons with the same capture date overlap each other in any way, correct?

OK, not sure what you have done to this point, but you can try what may be a simpler approach for you using 2 tools:

Summary Statistics
Dissolve

Here's the idea:
- If you run Summary Statistics on your feature class that has an area field, by case (your capture date field), to sum area, you will get a table of total area of polygons per capture date.  This will be a sum total counting any areas overlapping.

- Then if you run Dissolve to produce a test feature class, dissolving by capture date, allowing multipart features in the result, you will get a new area field where any overlapping areas have been 'made 1'; i.e., the common area will only be computed once.  This means that for any set of 'capture date' features that has any overlap, the 'summary' area figure will not match the 'dissolve' area figure.

- In order to 'find' the problem features, then join the table result to the feature class result by your capture field and perform a selection query on the areas, basically dissolveArea <> summaryArea.

Get the picture?  Note that if you're not using a gdb feature class, you may have to re-calculate an area field in the Dissolve result - can't remember if you can execute a summary from within Dissolve.

If you need more help you can zip and attach a sample of your data - it really isn't a long process, so if you don't see what to do I can send you the results back with a better explanation.
0 Kudos