[R-sig-Geo] Kriging

[ONKELINX, Thierry]

Dear Robert,

In my opinion you need to look for some secundary information that will
allow you to split your data into subregions. Trying some arbitairy
dvisions and then choosing the one with the best fit may lead to an
overfit of the data.

If the phenomenom that you're studying depends on the geology then you
could look for a stratification by e.g. rocktype.

HTH,

Thierry

------------------------------------------------------------------------
----
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature
and Forest
Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
methodology and quality assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium
tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be
www.inbo.be 

Do not put your faith in what statistics say until you have carefully
considered what they do not say.  ~William W. Watt
A statistical analysis, properly conducted, is a delicate dissection of
uncertainties, a surgery of suppositions. ~M.J.Moroney

 

> -----Oorspronkelijk bericht-----
> Van: r-sig-geo-bounces at stat.math.ethz.ch 
> [mailto:r-sig-geo-bounces at stat.math.ethz.ch] Namens Robert Helber
> Verzonden: donderdag 26 juli 2007 22:25
> Aan: R geo
> Onderwerp: [R-sig-Geo] Kriging
> 
> I am attempting to krige a very large 2D region using the 
> geoR package (http://cran.ssds.ucdavis.edu/).  The geoR 
> kriging works well for regions that have stationary data.  My 
> big region has areas where the trend and possibly the 
> variogram changes.  Because of this I am trying to split up 
> the big region into smaller regions that have more stationary 
> data so that I can krige each region separately.  The problem 
> I'm having is that sub-regions that are too small have too 
> little data to make a good variogram.  I need to choose the 
> correct size sub-region. 
>   My total region is very big and need to automate the 
> process.  Is there R software out there that will help with 
> this?  Is there a way to choose the size of the subregion in 
> a systematic and logical way?
> 
> Thanks.
> 
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>