Jump to content

 
Photo

Spatial Statistics

- - - - -

  • Please log in to reply
2 replies to this topic

#1
spesseh

spesseh

    Key Contributor

  • Validated Member
  • PipPipPip
  • 62 posts
  • Norway

Hi again :)

I need some advise from some of the more knowledged people in here about some spatial statistics.
I have a GRID/FISHNET with polygons as my unit of observation. In these polygons i am gathering different spatial datasets and information for the underlying area.
What i want to do in this instance is to count the coverage of a raster inside the cells.
In basic i have my vector grid with the cells, and a raster layer containing the information i want to join into the cells.
Any suggestions on how to do this. Ive tried the extract by polygon tool in ArcGis, but it crashed. I am thinking it has some kind of limitation. In my instance i have aprox. 68000 polygon-cells.

So. How should i solve this?
Thanks!

Andreas
Andreas

Visit my Blog: http://www.gisintersect.com

#2
spesseh

spesseh

    Key Contributor

  • Validated Member
  • PipPipPip
  • 62 posts
  • Norway

Sorry. I was a bit quick to post the question.
I found this functions in both the GeoStatistical Analyst extension for ArcGis and the Hawths tool Package.
Andreas

Visit my Blog: http://www.gisintersect.com

#3
spesseh

spesseh

    Key Contributor

  • Validated Member
  • PipPipPip
  • 62 posts
  • Norway

I tried to do a thematic raster summary of a raster for all of my 68634 vector cells. The problem is that the Hawth Tools - Thematic Raster Summary did only summarized the values in the eastern hemisphere.
What could be the problem here? I tried to export the cells that did not get calculated, and did a summary only on that layer. Did not help. The western hemisphere is reluctant to any summarization.
See the image below to see what cells got a new calculated value (those selected). The other just contains NoDATA.
Posted Image
Andreas

Visit my Blog: http://www.gisintersect.com




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

-->