[STATSGRASS] Trend surface analysis issue - sending results back to GRASS

[Roger Bivand]

On Wed, 27 Sep 2006, Carlos "Guâno" Grohmann wrote:

> Hello all
> 
> Since the new sp and spGRASS6 packages, I haven't done any trand
> surface analysis yet, so I'm a little lost here..
> 
> in the old day, I would do this:
> 
> g1r<-surf.ls(1,gradiente$east,gradiente$north,gradiente$num1);
>  g1g<-trmat.G(G,g1r);
>  rast.put(G,lname="gradiente.trend1",layer=g1g);
> 
> now that trmat.G doesen't exists anymore, how do I do to send my trend
> sufaces back to GRASS?

Use the standard gstat() approach, with the degree= argument. The gstat 
package and surf.ls() in MASS use the same coordinate shrinking approach 
and give the same results.

Roger

> 
> thanks
> 
> Carlos
> 
> 

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no