slope.aspect solution (maybe)

[Grass package]

> A few weeks ago I posted a message requesting help with a problem I was having
> with r.slope.aspect.  When I used this command on a dem the resulting maps
> came out with linear areas of 0 slope in areas that should have contained 
> relatively steep slopes.  We'll, I think I found a solution (or at least a
> fudge).
> 
> I did a reclass of all the areas that had 0 slope and turned them into 
> areas of "no data."  I then used r.surf.idw2 to interpolate the values for all
> the areas of no data.  The resulting map seems much better than the original
> (although it is not perfect).  Does anyone know if I violated any cardinal
> rules by doing this?  If so, any other solutions?
> 
> Bill Green
> wg1938 at cnsunix.albany.edu
> 

Hi Bill,

You've hit upon a solution similar to the one I've used for a bad DEM.
However, if you do this on the Slope layer instead of on the DEM itself, I
think you are introducing errors. Try turning the DEM to 'no data' where
the Slope = 0, then re-interpolating the DEM and recalculating Slope from
that, and see if you end up with a better slope map. Of course, this
isn't going to be perfect either...

Martijn

P. Martijn van Leusen -------------------------------------------------------+
Leverhulme Research Fellow, The Wroxeter Hinterland Project                  |
University of Birmingham Field Archaeology Unit                              |
Edgbaston, Birmingham B15 2TT                                                |
Phone +44 121 414 5513  Fax +44 121 414 5516  E-mail P.M.van-Leusen at bham.ac.uk