[R-sig-Geo] RE : Variance unchanging with krige in gstat with iterations

[Tobin Cara]

Ah yes, thank you. Sorry for being new to this field. I appreciate your help.

________________________________________
De : Paul Hiemstra [p.hiemstra at geo.uu.nl]
Date d'envoi : jeudi, 12. novembre 2009 12:45
À : Tobin Cara
Cc : r-sig-geo at stat.math.ethz.ch
Objet : Re: [R-sig-Geo] Variance unchanging with krige in gstat with iterations

Hi Tobin,

This is to be expected. the kriging variance only depends on the
variogram and the distribution of the observations. If these don't
change, that the kriging variance remains constant. You can have a look
at the kriging variance equations in any textbook [1] or alternatively
in one of my papers [2]. An interesting discussion on this can be found
on the AI-Geostats mailing list [3].

cheers,
Paul

[1] e.g. @BOOK{Christensen1996,
  title = {Plane Answers to Complex Questions: The Theory of Linear Models},
  publisher = {Springer, New York},
  year = {1996},
  author = {Ronald Christensen},
  edition = {Second},
  note = {496p},
  owner = {hiemstra},
  timestamp = {2007.10.19}
}
[2] http://intamap.geo.uu.nl/~paul/files/PHiemstra_CandG_paper.pdf, eq (4)
[3] http://www.mail-archive.com/ai-geostats@jrc.it/msg02890.html

Tobin Cara wrote:
> Hello,
>
> I am performing kriging with an external drift and ordinary kriging in a loop to have predicted values at each time step.
>
> The predicted values are changing reasonably, however, the variance is not changing after the first time step.
>
> Here is my simple for loop:
>
> # KED
> list_ked <- vector("list",72)
>
> for(i in 1:72){
>  list_ked[[i]] <- krige(Prec00[,i]~Z, locations=DataCoord, newdata=elev, model = fitted_vario)
> }
>
> # OK
> list_ok <- vector("list",72)
>
> for(i in 1:72){
>      list_ok[[i]] <- krige(Prec00[,i]~1, DataCoord, newdata=coord_points, model = fitted_vario)
> }
>
> Would anyone know the solution?
>
> Thank you very much,
>
> Cara
>
>       [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>


--
Drs. Paul Hiemstra
Department of Physical Geography
Faculty of Geosciences
University of Utrecht
Heidelberglaan 2
P.O. Box 80.115
3508 TC Utrecht
Phone:  +3130 274 3113 Mon-Tue
Phone:  +3130 253 5773 Wed-Fri
http://intamap.geo.uu.nl/~paul