I have a raster image that has been reclassified into 15 equally incremented values/levels. This raster represents "sinkhole risk" in a certain county. So level 1 is the lowest risk region and level 15 is the highest risk region.

I also have the shapefile data for the actual sinkholes existing in that county.

My hope is to come up with some method of statistically analyzing the data by answering the following question:

"What percentage of the sinkholes are in the risk regions 1-3?" or

"What percentage of the sinkholes are in the risk regions 4-6?" etc...

Any suggestions on a tool to use or a general approach? I just need some guidance in the general right direction!

I posted a photo below for more clarity! The photo has been simplified into 3 values/levels, but my real map has 15 values.

Photo Explanation

I am using ArcMap 10.1.

migrated from stackoverflow.com Jul 8 '14 at 10:04

This question came from our site for professional and enthusiast programmers.

  • Cutest image on gis.se so far imho :D – Alex Tereshenkov Jul 8 '14 at 10:08

You can approach it in different ways, better working with same type of layers: both raster or vector files. I would do in a quick&dirty way:

  • Use "Polygon to Raster" to convert sinkhole shapefile into a raster: Value field "FID", Cell assignement "Maximum Area", Cellsize same of risk raster, on Environments/Extent snapped with risk raster.
  • Reclassify all sinkhole values to 0 (or with Raster Calculator use the expression: Con(sinkhole >= 0, 0, sinkhole)
  • Sum the risk and sinkhole raster with Raster Calculator
  • Find the list of levels and the number of pixels covered on the DBF table of the output.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy