Given the DEM of a certain Country, and a fishnet that is superimposed on it, I need to compute the average elevation and slope corresponding to each square in the fishnet.

[I am using ArcGIS 10.3.1, licence type: Advanced.]

These are the DEM (shades of grey) and the fishnet (red): enter image description here

The DEM is not a unique raster, but a handful of smaller maps. Their resolution is 10x10 km, whereas the fishnet squares are 2x2 km.

I tried to create a Mosaic Dataset, and then I used Extract -> Clip from ArcToolbox using the DEM as input features and the fishnet as clip features. I thought this wouldhave produced a unique raster map, but I was wrong, and the result is essentially a grid with the DEM resolution (10x10 km) and a bunch of table attributes which do not refer to elevation. enter image description here

The intended outcome was, instead, to get the average elevation and slope for each one of the smaller tiles in the fishnet.

What is it the best strategy to achieve this?

  • 1
    Possibly a duplicate, but there is the additional question here of handling multiple DEMs.
    – Tangnar
    Commented Mar 6, 2017 at 13:40

1 Answer 1


I think there could be multiple ways to approach this. I would do this:

  1. Combine your multiple rasters into a single raster with Mosaic To New Raster tool, which is different from Mosaic Dataset.
  2. Create a Slope grid from the raster mosaic.
  3. Use the Zonal Tools to calculate your average elevation (from DEM) and slope (from Slope grid) for each cell of the fishnet. I think the Zonal Statistics as Table would give you what you want.

The Zonal Statistics as Table will iterate and calculate the statistics you are interested in for each box in the fishnet if you give it a unique ID - like OBJECTID, though there is a warning in the tool's instructions about using this reserved field name, and a link to instructions for creating a new field and making your own unique ID - for the 'zone field'. You'd do it twice, once on with the DEM, and once with the slope.

The output will be tables with the unique ID and the stats you want, which you can then join back to the fishnet layer.

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