DEM Reconditioning does not change DEM

Question

I'm using ArcGIS Pro version 3.5.2, standard license with 3D and spatial analyst extensions. I am using DEM Reconditioning tool to burn my stream into the DEM. The tool finished successfully but the resulting AGREEDEM is essentially the same as the input DEM (the difference is in the 5th decimal places). I have look through the forum and find this thread:

Solved: Problems with Burn Stream slope and DEM Reconditio... - Esri Community

I have made sure my X, Y, coordinate system and linear unit of everything (project, map, raster, geoprocessing environment and everywhere else I can find) are in meters and it did not help.

I tried to increase the z unit in the DEM reconditioning tools A LOT (in the order of millions) to force the burn to be more significant but the results is the AGREEDEM's value are all round numbers and the change in values are large discrete values instead of gradual change.

Has anyone else encountered this and found the solution?

MarkBoucher · Answer

I have a lot of experience with the ArcMap version of Arc Hydro and some with the ArcPro version.

Some ideas...

* Make sure you confirm the processing extents cover your project. In the ArcMap version the processing extent would sometimes get changed for some reason I could never figure out. For example, the processing extent would be reset to match the limis of the batchpoints. A confusing situation I had to deal with.
* Make sure all layers (DEM and agreestream) are in the same coordinate projection.
* If the values of the AGREEDEM are "all round numbers," I assume you mean they are integers (e.g., 234) and not floating-point numbers (e.g., 234.56). This can happen during the "burning" process.

The most likely cause is that one of the geoprocessing tools you used to subtract values from the RAWDEM is outputting an integer raster instead of a floating-point one. Check the output data type settings for the specific tool you used, and also verify the data type of your original RAWDEM. You want to ensure the final output retains the real number precision of your original data. (Some of this answer I got from Gemini [Googles AI])