I'm doing urban change prediction in Land Change Modeler (TerrSet software), and i can't run Transition Sub-Model.

I don't know why, but I think it is because I have nodata value in Variable (slope image, elevation image, distance to roads) and I don't know how I can remove these values.

I tried in ArcGIS (reclass, copy raster).

Nodata value is -3,402823e+038

  • 1
    Does your raster have NoData? you have a NoData value set but do any pixels actually have that value? Do you have cells with a different value you want to be NoData? What is the raster type (GRID, Tiff, img etc..) and pixel type of your raster? Are there any specific error messages? A rough stab-in-the-dark: could be because your raster is floating point and the model expects classifed and attributed integer raster. Aug 27 '15 at 4:00
  • Raster is img. type, pixel type is floating , and I must convert to raster format (rst.) in TerrSet...error message is "Error in command line"....when I show nodata value in ArcGis in red color, data(raster's) frame shows in red...i tried unsinged integer of pixel value, but I have same error message
    – Alma
    Aug 27 '15 at 22:59

There is a useful method using Con and IsNull statements in the raster calculator to convert NoData to a different value.

HowTo: Convert NoData values to other values for raster data

enter image description here

  • I can vouch that this method works as I have used it many times.
    – TsvGis
    Aug 27 '15 at 6:49
  • I tried but I still have nodata value (now is -32768)
    – Alma
    Aug 27 '15 at 23:21

@Aaron's answer might be the best solution for general raster files (thematic or continuous data). If you have a thematic raster you could also use the Reclassify tool (see below). In the example NoData will be replaced with 0.

enter image description here


Apply Abs tool with raster calculator

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.