[GRASS-user] cost analysis

[Dylan Beaudette]

Maria,

A simple approach would be to construct a composite "travel friction" map. I 
have done this before (example to follow) using a slope map, wooded areas 
map, and water feature map. This was for predicting least cost trails through 
a wilderness area minimizing:
-travel across high slope gradient
-travel through water features

and maximizing:
-travel through wooded areas

The details can be found here:
http://casoilresource.lawr.ucdavis.edu/drupal/node/244

The general concept is:
create a slope map, as as starting "friction" surface. 
encode other information into this map via r.mapcalc
run r.cost
run r.drain

Cheers,

Dylan

On Thursday 29 June 2006 02:47, María Piquer Rodríguez wrote:
> Hello,
>   I would like to perform a nature connectivity analysis, generation of
> paths connecting natural areas, taking into account several cost grids (eg:
> land-use, level of protection, kind of soil,vegetation,etc).
>
>   I have read that the r.cost is an option
> (http://grass.itc.it/grass61/manuals/html61_user/r.cost.html) but I would
> need to use more than one cost grid. The option that r.walk
> (http://grass.itc.it/grass61/manuals/html61_user/r.walk.html) offers is not
> so suitable for me as the elevation in my study is not so relevant (it is
> quite a flat area).
>
>   Does anyone knows if with the r.cost it is possible to integrate more
> than one grid for the least cost computation? or if for r.walk the
> elevation grid can be another grid(s) instead of the elevation one?.
>
>   Of course the option of computing, for several overlaying grids, the per
> cel cost  in the r.mapcalc before hand is always possible.But I was
> wondering if there is an application available that performs such a
> multigrid least-cost calculation automatically.
>
>   The option that r.drain
> (http://grass.itc.it/grass61/manuals/html61_user/r.drain.html) gives is
> also very interesting for my study. The generation of the least cost path
> is a very good way to select connectivity areas between natural patches.
> But again r.drain operates taking into account an elevation data set and a
> cost grid one. Does it exists a module as r.drain taking into account
> several cost grids, not especially the elevation one, for calculating a
> least-cost path?
>
>   Anyother option that I did not mention here is more than welcomed!
>
>   Thanks you very much for your help.
>
>   Regards,
>   María Piquer
>
>
>     María Piquer Rodríguez
> Mozartstr, 58
>   53115 Bonn
>   Germany
>
>
>
>
>
>
>
> ---------------------------------
>
> LLama Gratis a cualquier PC del Mundo.
> Llamadas a fijos y móviles desde 1 céntimo por minuto.
> http://es.voice.yahoo.com

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