[R-sig-Geo] block and output grid size in block kriging

Edzer Pebesma edzer.pebesma at uni-muenster.de
Thu Feb 12 16:30:23 CET 2015



On 02/12/2015 04:19 PM, subash wrote:
> Dear Edzer, 
> 
> To understand why block kriging results vary,with different grid sizes, to
> keep gridsize identical to grid cells, and smaller than grid cells.
> 
> library(gstat)
> data(meuse)
> coordinates(meuse) = ~x+y
> data(meuse.grid)
> gridded(meuse.grid) = ~x+y
> 
> lzn.vgm<- variogram(log(zinc)~1, meuse)
> lzn.fit1 <- fit.variogram(lzn.vgm, model=vgm(psill=1, model="Sph",
> range=900, nugget=1))
> plot(lzn.vgm, lzn.fit1)
> 
> # Ordinary kriging
> lzn.ok <- krige(log(zinc)~1, meuse, meuse.grid, model = lzn.fit1)
> 
> #Block kriging 
> lzn.bok <- krige(log(zinc)~1, meuse, meuse.grid, model = lzn.fit1,
> block=c(40,40))
> 
> # Choosing a random location with the centre coordiantes(xx,yy) and find the
> corresponding block kriged estimate
> xx = 181140
> yy = 333100
> est.bk = lzn.bok$var1.pred[(lzn.bok$x==xx)&(lzn.bok$y==yy)]
> [1] 6.499158
> est.ok =lzn.ok$var1.pred[(lzn.ok$x==xx)&(lzn.ok$y==yy)]
> [1] 6.499624
> 
> Why are the kriging estimates at a point different in block and ordinary
> kriging, when the grid size is isentical to (40 x 40) or less than(20x20)
> the meuse data.

because lzn.ok contains ordinary point kriging estimates, for points
laid out on a grid, and lzn.bok contains ordinary 40 m x 40 m block
kriging estimates: a different quantity is predicted.

> 
> Thanks
> 
> 
> 
> 
> 
> 
> 
> diff= m.bk - est.bk
> 
> 
> 
> --
> View this message in context: http://r-sig-geo.2731867.n2.nabble.com/block-and-output-grid-size-in-block-kriging-tp7587770p7587780.html
> Sent from the R-sig-geo mailing list archive at Nabble.com.
> 
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
> 

-- 
Edzer Pebesma
Institute for Geoinformatics, University of Münster,
Heisenbergstraße 2, 48149 Münster, Germany; +49 251 83 33081
Journal of Statistical Software:   http://www.jstatsoft.org/
Computers & Geosciences: http://elsevier.com/locate/inca/398

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: OpenPGP digital signature
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20150212/d43ed3a4/attachment.bin>


More information about the R-sig-Geo mailing list