3

I have about 300 MODIS satellite with different acqusition dates, covering the same area. I am not used to work with so many tiles at once so I ask for advice. I am working with ArcMap10.1

What I want to do is:

  1. Stack all the images on top of each other
  2. clip them with the same extent.
  3. Run raster calculator for each of them, with the same function.

I guess my biggest problem is how to stack them in a good way so I later can run my tools on the whole pack.

Best regards!

  • 1
    It sounds like you are proposing creating a data cube (en.wikipedia.org/wiki/Data_cube), which I think can be constructed in remote sensing software such as ERDAS. However, you could also achieve your desired result using model builder or python to automate the clipping and raster calculation of each file in turn. – sgrieve Oct 24 '12 at 12:48
1

As @sgrieve mentioned, the best solution using ArcMap 10.1 is to build an iterative model (see attached). In case you are unfamiliar with model builder:

  • The first input "default.gdb" is a workspace variable that I added.
  • If you want to clip the rasters to the polygon geometry rather than the extent, make sure to select that check box in the Clip tool.
  • If you want to create a separate clipped raster, use the %Name% and a identifier such as "_clp" (e.g. "Name"_clp) method. If you are not interested in creating intermediate data, select "Delete Intermediate Data" within the model builder or assign a single name (e.g. Raster) to the clip output and specify overwrite options in Geoprocessing > Geoprocessing Options.
  • Finally, the output from the raster calculator should have a unique name, so add to the output path something like C:\Temp.gdb\%Name%_calc.

A python solution would be very easy too--let me know if you would prefer to go that route. Good luck.

enter image description here

  • Thanks alot for your answer! That was really detailed. I have never used Iterations before, seem to be very usefull. My project took another turn and never needed to use it. I processed the data with another program to get the informationen I intended to find, which reduced my data set to an amount not needed to go trough a model. – user10685 Nov 10 '12 at 10:48

Your Answer

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