1

enter image description hereI have two data layers of the same geographic area, one which has a much coarser resolution and the other that is a much finer resolution. As such, the first layer has some cells that do not overlay perfectly with the second layer. That is to say, that some cells of the first, more coarse layer, are not completely aligned with the edge of the second layer.

I need to remove the cells from the coarse raster layer that are not fully covered by the finer one. No cells lie completely outside the second layer, but a few are partially uncovered, and I want to remove them from the raster and create an output that doesn't have these cells. Essentially, I want to edit the first layer so that every single cell lies completely within the coverage of the second layer. I want to remove the grey cells in the image that are peeking out from behind the landcover data in the foreground.

I've tried extract by mask, but as best I can figure, it's not working because the center of the cells still lie inside the polygon.

Does anyone know how to alter the environment somehow to make the extract by mask exclude cells that lie even partially outside the polygon? Or am I using a completely wrong method?

1 Answer 1

1

You may need to use Resample tool to change the spatial resolution of the coarse raster image by decreasing the pixel size and make the resolution similar to the fine image. After changing the spatial resolution, you can run extract by mask to include more pixels with the mask polygon. However, the image size of the coarse resolution image will increase, which may affect the efficiency and increase the processing time, so you have to be careful about it.

1
  • I thought about that, but my problem is the coarse raster contains spatially sensitive data, so if I decreased the pixel size it wouldn't correspond to the overlain map in the way I need it to. Do you know of a way to just go and manually delete individual cells from a raster?
    – Cb14
    Commented Jan 9, 2018 at 22:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.