I am using National Land Coverage Data (NLCD) for a project. I am trying to divide a feature into separated features via a grid system so that I can alter and calculated the different land coverage per grid cell. I have created a ten by ten fishnet, but I have been unable to divide the NLCD into separate feature either as rasters or polygon shapefile. I've used the split function, but it did not keep the extent nor the orientation of the grids. Also, I tried intersecting but that didn't work either. What would be the best way to accomplish this? 10 by 10 gridNLCD data

  • 1
    What about Union? It should combine all attributes into a single layer. – klewis Nov 9 '17 at 22:45
  • Union is a good place to start if you want to do this interactively; how good is your python? I use a script to extract for each feature in a grid which works quite well and would take little modification for your use.. Are you extracting shapefiles in a folder, feature classes in a geodatabase or every layer in a map? – Michael Stimson Nov 9 '17 at 23:05

Option 1:

Provided both the land classes and your fishnet are polygon feature classes, then, as others mentioned in the comments, it appears that the tool you want is Union. As per the documentation, the tool:

Computes a geometric union of the input features. All features and their attributes will be written to the output feature class.

You will then have a file which contains all the attributes of your land classes and all the attributes of your fishnet.

You can then use tools like Summary Statistics to calculate percentages or totals of different land uses per grid cell.

Option 2:

If your land classes are a raster layer rather than vector, then I suggest using zonal statistics. You'll need the Spatial Analyst extension, however. The tool

Calculates statistics on values of a raster within the zones of another dataset.

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.