topic Re: From contour lines to slope classes in ArcGIS Spatial Analyst Questions

From contour lines to slope classes

MaikeHeuner — Sun, 01 Jun 2014 10:12:13 GMT

Hi,

contour lines describe the steepness of the slope. I would like to define slope classes (e.g. gentle or steep) from the separation distances of contour lines e.g if the separation distance is greater than 100 m, than the slope will be gentle. If the separation distance is smaller than 100 m, than the slope will be steep. I would be happy if anybody can explain me how to do it in ArcGIS Desktop.

Thank you

Re: From contour lines to slope classes

SergeyChernyshov — Mon, 02 Jun 2014 13:09:50 GMT

If you have source raster data of contour lines, you can use Spatial Analist -> Surface -> Slope. The result is gradient of surface for each cell of a raster.

Re: From contour lines to slope classes

curtvprice — Mon, 02 Jun 2014 16:46:19 GMT

I would like to define slope classes (e.g. gentle or steep) from the separation distances of contour lines e.g if the separation distance is greater than 100 m, than the slope will be gentle. If the separation distance is smaller than 100 m, than the slope will be steep.

I can't think of an easy way to accomplish the analysis you propose, but I think a better approach would be to use the Topo To Raster tool to create an elevation raster -- then use the Slope tool to calculate slope for each cell from the elevation raster. From there you can reclassify the slope values using the Reclassify or Slice tools to assign a slope class to each cell.

Re: From contour lines to slope classes

MaikeHeuner — Tue, 03 Jun 2014 08:03:21 GMT

Thank you for your suggestions, but these usual ways don't work here on tidal flats (see attached) with low slopes. The contour lines make differences vizualising the river direction, but the method surface-slope not really. With this method I can't consider the direction. Also reclassification doesn't bring a benefit.

Re: From contour lines to slope classes

SteveLynch — Tue, 03 Jun 2014 16:39:44 GMT

Try PathDistance with only the following inputs
- Input raster or feature source data = your contour feature class
- in the Environment -> Raster Analysis -> Cell size = cell size of your DEM or experiment with different values

-Steve

Re: From contour lines to slope classes

MaikeHeuner — Wed, 04 Jun 2014 05:52:48 GMT

Thanks a lot, that will work with some further steps of reclassifing, converting to polygon features, elminating small areas and so on...

Re: From contour lines to slope classes

curtvprice — Wed, 04 Jun 2014 17:40:26 GMT

Straight euclidean distance from your contours may work too:

1. convert contours to raster
2. clean up garbage using Generalization toolset
3. Euclidean distance (source - contour cells)
4. Select cells where focalmax = value (larger window) (use SetNull)
5. maybe convert to points for manual cleanup
6. Convert distance values to integer and expand using nibble - reclassify result

Re: From contour lines to slope classes

GerryGabrisch — Tue, 10 Jun 2014 19:48:50 GMT

Thank you for your suggestions, but these usual ways don't work here on tidal flats (see attached) with low slopes. The contour lines make differences vizualising the river direction, but the method surface-slope not really. With this method I can't consider the direction. Also reclassification doesn't bring a benefit.

Low slopes shouldn't make a difference. Convert the contours to a surface model using the advice posted by others like topo to raster. Generate a slope raster from the surface model. Use focal statistics in the Spatial Analyst toolbox to return a new surface with the average slope over some user-defined area. Reclassify the focal statistics output.

It also sounds like you want to view the direction of flow. Is that right? If you want to see which way the water would flow across your surface then generate a flow accumulation surface from your surface model using the Spatial analyst /hydrology tools. After you have created the flow accumulation surface, go into the flow accumulation symbology, change the gamma stretch to something like 20 and set the standard deviation to 0.5. This will really bring out the flow directions.
I hope this helps.