Suggestion request

[DavideGalimberti]

Good morning!

I am quite new to ArcGIS. Here is my problem: I have created a DTM and its relative "Slope map" (using the "slope" command from the Spatial analyst tool).

I would like to create a slope variation map now, starting from this "slope map", in order to detect where there is a variation larger than 10°(which means that sudden and big variations can be found there).

I have tried this “block statistics” command, calculating the Standard Deviation of 3x3 cells blocks. I have tried also to calculate the “range” of these blocks but the result differs a lot from the previous one, of course. My question is: are there other commands or other ways to calculate this slope variation? And, in general, are there other commands with which it is possible to make calculation and operations on the cells belonging to a raster?

Thanks in advance

[JayantaPoddar]

Hi Davide,

You could use Reclassify (Spatial Analyst) tool to categorize the slope raster.

e.g.

Slope          Reclass_Code

0-10                      1

10-90                    2

So all the rasters with value "2" would have a variation of 10° or more.

You can choose the slope range and no. of classes, as desired.

[JayantaPoddar]

You can also use Raster Calculator (Spatial Analyst)​, with similar expression

Con("SlopeRas"<10,1,2)

where "SlopeRas" is the name of the Slope Raster (DTM)

1 is the value that is defined to the pixels with Slope lesser than 10°.

2 is the value that is defined to the pixels with Slope more than 10°.

[DavideGalimberti]

Thank you very much. I didn't know about this "reclassify" command.

Btw, this actually doesn't solve my problem: I want to identify the slope variations between neighboring cells (for example if there’s a pixel with a slope=45° and the adjacent one with a slope=60°, I would like to detect them), Is there a way to do this?

[JayantaPoddar]

Not sure if it will work for you.

While creating the Slope Raster (DTM), you could choose the output measurement as "Percent_Rise" than "Degree".

[DavideGalimberti]

I already tried to do that, but the result is like in the slope raster: the only difference is that values are in percent and not in degrees

[SepheFox · Accepted solution]

Hi Davide, I think what you're looking for is FOCAL Statistics: ArcGIS Help (10.2, 10.2.1, and 10.2.2).

[DavideGalimberti]

Hi Sephe! Thank you! I think it should work, I will try to use that command!