1

currently I am dealing with the creation of a landuse file. I have geoprocessed a google earth .tiff-file over my DEM and created three polygon featureclasses via the editor tool containing areas of a certain landuse. After the conversion into raster files of course the number of rows and columns is different to the ones in my DEM - is there a way to fix that? (tried clipping already, no success)

In the end I'll need one single landuse file with the different categories (probably there's a easier way to achieve that!?) I am using arcgis 10.1

  • I know that the subject is pretty old, but have you managed to find the solution? I have exactly the same problem (also landuse files). I tried couple of things none of them seem to work. – MariaMaria May 1 '17 at 15:28
3

ArcGIS! Happy to hear that jaja I am in troubles with qGIS but you could try to cut your rasters (I understand your have 3) using a polygon (extract by mask) and then use Resample to change the pixel size, it should work. There is another software called Imagine from ERDAS I had heard that it does what you try to do! luck !

  • toolset clip (raster), which also clips the input raster with a feature-file, only generates strange data -> the input DEM-file for a certain region has 132 columns and 84 rows, whereas the clipped version only has 90/49... it doesn't seem to make any difference wether i use my DEM or any other raster file (which has the same number of columns and rows) to clip from - the output always has 90/49 instead of the original 132/84. so it does really seem to clip my file- thus the change in cols/rows!? i tried the EXTRACT BY MASK tool but couldn't get it to work ("error during conversion") – cataloop Jul 7 '15 at 16:10
  • If you are using arcGIS does not matter the number of rows and columns, but you need to have the same pixel size, and you could repair it resampling your data, extract by mask should work if you have a raster and a polygon in the same projection (WGS 84 for example), check it some times the software display them in the same place but they are not in the same projection. – Nelly Jul 9 '15 at 13:48
-2

Using Extract by Mask worked for me after resampling SEVERAL times. My purpose was to turn a vector into a raster to have the same rows/columns as other rasters. To do this I took the following steps:

  1. Converted the vector into a raster.
  2. Used Raster to Domain to create a bounding box of the extents of the ideal raster, creating a polygon boundary line.
  3. Clipped the converted raster to this bounding box.
  4. Resampled to the ideal cell size bi-linearly, and using processing extent and snap raster to the ideal raster.
  5. Projected the raster to the WEB MERCATOR AUXILLARY SPHERE. I had trouble finding out that I needed this projection rather than WGS 1984. Also when projecting and doing these processes, keep snapping the raster and changing the processing extent accordingly.
  6. Reclass accordingly and THEN extract by mask.

And finally the rows and columns and extent are the same!

  • 1
    I saw that your answer is getting downvoted. You can improve future posts by using formatting to make them more readable. For example, see how much easier this answer is to read now that I converted it from a long, run-on paragraph to a numbered list. You could improve it further by adding screenshots for the more complicated steps. – csk Mar 6 at 19:58

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.