0

I am trying to use Zonal Statistics to determine the highest point within polygons. I previously did this in ArcMap w/o problems, but now I am finding that certain polygons are only partially digitized or not at all. If I do multiple runs of my python program, it works sometimes, partially others, sometimes not at all. If I execute the command individually through the Geoprocessing menu, I get the message "WARNING 010566: Some zones may not have been rasterized." I have consulted ESRI help on this issue and also other questions related to this command, but find my situation is different. (I am not including the code because the Spatial Analyst tool is also failing in similar ways.)

I am working with a .5m resolution raster. That is also the pixel size of the output raster. Both have the same projection (WGS 1984 UTM 16N). The polygons are much larger than the pixel size, so this is not the problem mentioned in the ESRI help file. I am attaching an illustration: the red outlines are the polygons, the white/gray/black filling is the result of zonal stats. As can be seen, several are not digitized and the one in the bottom center only partially (others are cut off like this along a horizontal line in other sections of the image). Any suggestions? enter image description here

1
  • You have not presented any code so I have removed your tag for ArcPy.
    – PolyGeo
    Commented May 5 at 20:18

2 Answers 2

1

This is where I would start looking:

  1. Looks like the processing extent was smaller than the extent of the polygon feature class?
  2. What other properties are you setting? Cell size, mask?
  3. How were you running your python script, in and IDE or directly within ArcPro where it might have inherited some global setting?
  4. Finally have you run check geometry on your polygons, may be they are malformed in some way, such as incorrect ring order?
2
  • Thanks for your thoughts. I don't think it is a question of extents, since some of the empty polygons are within a box defined by filled polygons. Geometry is a good suggestion, but I checked and the polygon file is OK. Running the program as a script from within an ArcPro tool (and also, as mentioned, using just the lone tool). Not setting any parameters. The polygons are created from a previous raster using raster to feature tool. As mentioned the cell size stays constant.
    – William
    Commented May 6 at 0:20
  • Hmm interesting, next thing is to share this data or a sample of it to see if others can replicate it?
    – Hornbydd
    Commented May 6 at 9:26
1

Even though the polygon geometries checked out OK, it must be a problem with them. I exported the polygons to a new feature class, reimported it and the zonal stats appear to work. My concern is that the polygons were created by an ESRI raster-to-polygon routine, so that doesn't inspire a lot of confidence.

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.