topic Re: Minimum slope distance in ArcGIS Spatial Analyst Questions

Minimum slope distance

SteveMundall — Mon, 31 Aug 2015 03:42:51 GMT

Hello,

I am trying to define areas of slope greater than 40% that have a horizontal displacement greater than 25 ft.

I haven't found a way to set a vertical or horizontal limit with slope calculation and am hoping for some help to accomplish this.

If someone can give me some pointers on how this might be done I would be grateful!

Arcmap 10.2

Thanks!

Re: Minimum slope distance

NeilAyres — Mon, 31 Aug 2015 06:14:31 GMT

There are many ways to process rasters in the SA tool set.

But first, what is the pixel size of your dem / slope, especially relative to your distance of 25ft.

You could segment your slope raster ie split out only those parts which have a slope greater than 40%.

Then perhaps look at Region Group & Tabulate Area tools.

Re: Minimum slope distance

DanPatterson_Retired — Mon, 31 Aug 2015 08:03:54 GMT

Neil is correct... given

Slopes > 40%
horizontal displacement greater than 25'
slope help
how slope works ...

http://desktop.arcgis.com/en/desktop/latest/tools/spatial-analyst-toolbox/slope.htm

http://desktop.arcgis.com/en/desktop/latest/tools/spatial-analyst-toolbox/how-slope-works.htm

This means that you could have a single cell that meets those criteria ... depending upon the cell size. If you got lucky, you could have larger areas of uniform slope > 40% but that may not be enough.

I would be tempted to say calculate the range using FocalStatistics

http://desktop.arcgis.com/en/desktop/latest/tools/spatial-analyst-toolbox/focal-statistics.htm

however, that just gives you the range in elevation over the neighborhood and not with respect to the center cell...unless of course, you want to run this twice, using a rectangular neighborhood of 1x2 and 2x1 and other kernels representing the 4 diagonals. You would then end up with the range of elevation differences with respect to the center cell to see if any meet your requirement.

So in summary if you have large cell sizes with respect to the window over which slope is determined (and the method)...forget it.

If the cell size is small, relative to the slope window...you might want to fine-tune your requirements and decide on what might constitute the best approach.

Re: Minimum slope distance

DanPatterson_Retired — Wed, 16 Sep 2015 02:06:56 GMT

did you resolve this?

Re: Minimum slope distance

SteveMundall — Wed, 07 Oct 2015 21:45:24 GMT

Thanks much for the tips!

Cell size is 1ft so fairly straightforward.

I also got some hints elsewhere that led me to perform following general steps.

Generate surface/slope raster
Generate aggregated slope raster to give general flow direction arrows
Generate surface/aspect raster from aggregated raster, then changed symbol to a flow arrow that rotates to direction of slope. This allows for easy indication of slope direction for later tracing.
Create reclass raster of steep slopes over 40% so that I can create a mask of only steep slope areas
Use the reclass to extract a mask of the surface
Generate Hydrology/flow direction and flow length from the masked surface

Now I had a raster of flow length only of steep slopes which I shaded with desired break values. Flow length is calculated along the slope starting from zero and counts up with each cell. This is fine but I need to define the whole region of the slope whenever the length is over the 25ft. Being this close, and since the area was limited enough, I just traced polygons over the appropriate regions using the flow direction arrows as a guide to defining the edges of the slope.

Perhaps you have better ideas of how to arrive at the end result like maybe a mathematical expression that would weed out the steep slope areas that didn't have the threshold length, but this worked well enough for me.

Re: Minimum slope distance

DanPatterson_Retired — Wed, 07 Oct 2015 23:17:38 GMT

sounds good...the extra effort to fully delineate automatically wouldn't be worth it unless you untended to do this many, many times.