I have a single band raster with values 1-9 (9 being the best and a cell size of 5m) this raster has been clipped to a feature class that has approximately 80 individual buffered areas of 250m (none of these buffered areas are spatially joined). Within each of the buffered areas of the raster, there are multiple cells with different values.

I need a single point within each buffered area that has the best value, but for the largest concentration of those values. If there is no value of 9 then a single point for a value of 8 and so on.

So I will end up with 80 points that indicate the highest concentration of the best value available. Just to be clear the point won't necessarily be in the middle of the buffered area, but somewhere within it.

How do I go about modeling this or a python script in ArcGIS Desktop 10.3.1?

  • 2
    What software (arc*, qgis, etc) and or libraries (eg pysal, arcpy, pyqis) are you working with? – RoperMaps Jun 22 '17 at 8:46
  • Sorry, ArcGIS 10.3.1 – Tigger77777 Jun 22 '17 at 9:04

I would suggest the following workflow:

  1. Zonal Statistics as table with your polygon feature class as zones, selecting MAXIMUM as your statistics type.

  2. Feature to Point on your polygons.

  3. Join table from 1. to points from 2.

This could be implemented in python, model builder or probably simplest just by clicking through.

  • I tried your suggestion but this didn't quite work, I have a workaround using a sort, iterators, feature to polygon, plus some other tools. It's not pretty and not exactly what I was after, but it works for now. – Tigger77777 Jun 23 '17 at 10:23
  • @Tigger77777 Glad you got something that works. Sounds like you used model builder. Why not upload a screenshot as an answer. – RoperMaps Jun 23 '17 at 10:34

This is what I came up with: The overall model creates the Weighted overlay and a FC for the points to be appended into, this data feed into the sub models one for each value (9 to 1). The sub models iterate over the buffered areas and creates a point for the largest area within each of the values.

Overall model

Sub model there are 9 of these

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.