[GRASS-user] real distance analysis

[Dylan Beaudette]

On Friday 02 June 2006 10:30, Michael Barton wrote:
> Hamish’s reply is the closest to what  you want I think. However, to
> actually derive a buffer to define equal ground distance around a point, I
> think  you might need a bit more explanation.
>
> First create a slope map of your dem (r.slope.aspect). Then, as Hamish
> says, run r.cost. This will calcuate a ‘cost’ based on the slope of the
> map. He implies that the cost value in the cost map = the ground distance
> to traverse each pixel (i.e., higher on steep slopes where the ground
> distance is greater and less on flat terrain where the ground distance is
> lower). I don’t know if this is how the algorithm works or not (or also
> includes other factors), but the references he lists will clarify. If not,
> you can use trig in the mapcalculator to figure ground distance using—as he
> says—the resolution of the map (i.e., the horizontal distance to traverse
> each cell) and the slope of the cell.
>
> Assuming that r.cost on a slope map * resolution = ground distance. You can
> use the resultant map to create a buffer around a point that represents
> equal ground distance in all directions.
>
> Michael
> __________________________________________
> Michael Barton, Professor of Anthropology
> School of Human Evolution & Social Change
> Center for Social Dynamics and Complexity
> Arizona State University
>
> phone: 480-965-6213
> fax: 480-965-7671
> www: http://www.public.asu.edu/~cmbarton
>
>
>
> From: Bödőcs András <bodocs at ludens.elte.hu>
> Reply-To: <bodocs at ludens.elte.hu>
> Date: Thu, 1 Jun 2006 22:26:32 +0200
> To: <grassuser at grass.itc.it>
> Subject: [GRASS-user] real distance analysis
>
> Dear All,
>
> I am trying to solve the following problem, maybe does somebody an idea and
> can help me…. (thanx in advance)
>
> I would like to examine from an exact point that territory, which is inside
> an determinated distance from the exact point. BUT I don’t want to use the
> map projected distance (=Buffer zone), but the datas of a DEM.
> I would like to measure the „really” surface, the climbing etc. For example
>  I need the maximum distance „cirlce”  (30km or 30 mph doeas not matter)
> around a point, but measured in the „filed”(=DEM) not on the map
> (=projected area).
> Does somebody have an idea, who can I start with it?
>
> Many thanx
> Regards
> Andras
>
> Bödőcs András
> ELTE Institute of Archaeology
> 1088 Budapest, Múzeum krt 4/B.
> Hungary
> T: +36-1-411-65-54,4855200/2916
> F: +36-1-411-65-53

An example calculation (minus the multiplication by resolution) of distance 
from trail to points of interest, via a slope cost map here:

http://169.237.35.250/~dylan/temp/r.cost_example.png

This will be part of the GRASS talk at the Where 2.0 conference, I will be 
sure to post a link to the finished presentation.

Cheers,

-- 
Dylan Beaudette
Soils and Biogeochemistry Graduate Group
University of California at Davis
530.754.7341