[R-sig-Geo] 3D kriging with gstat

Edzer J. Pebesma e.pebesma at geog.uu.nl
Tue Jan 4 14:31:14 CET 2005


Promodos, do you want to interpolate at each (x,y) location over a range
of z values? If yes, that is indeed 3D kriging.

If you want to make one single kriging prediction for each (x,y) 
location, but take the altitude of that location into account,
I would use the altitude as an external drift (or regressor)
variable. This is 2D universal kriging, by some called
external drift kriging. It can be done by the gstat package,
and several others, among which geoR.
--
Edzer

Prodromos Zanis wrote:
> Dear all,
> 
>  I want to do kriging with gstat for x,y,z coordinates (latitude, longtitude 
> and altitude) to interpolate irregularly distributed temperature station data 
> in order to produce a gridded dataset altidude corrected.
> 
> where can I find hints or examples on variogram modeling and kriging in 3D with 
> gstat?
> 
> I look forward for you help. 
> 
> thanks and regards,
> Prodromos Zanis
> 

-- 
Edzer




More information about the R-sig-Geo mailing list