Gstat for R: Block Kriging

[Nick Hamm]

Hi,

I've been using gstat under R and trying to do some
block--kriging.  I then want display the output from the block kriging
using the levelplot command.   However, I don't quite understand the
syntax heere

For eg. if I take the sample data--set that is provided and have a go at
one of the examples:

x <- krige(log(zinc) ~ 1, ~x + y, meuse, meuse.grid, 
     model = v, nmax = 40, block = c(40, 40))  

I get a data.frame that is the same length as if I did:

x <- krige(log(zinc) ~ 1, ~x + y, meuse, meuse.grid,
     model = v, nmax = 40)

So I'm not quite sure what output in the former means and how I can plot
an image of the block--kriging results.

If anybody has any suggestions, I would be grateful.

Nick



On Wed, 12 Mar 2003, Edzer J. Pebesma wrote:

> The majority of the functionality present in the gstat stand-alone
> program (http://www.gstat.org/) is now available as a package/library for
> the S language (R, S-Plus), again called gstat. The package provides
> multivariable geostatistical modelling, prediction and simulation, as
> well as several visualisation functions.  Gstat was started 10 years
> ago and was released under the GPL in 1996; the original stand-alone
> program is closely linked to several GIS systems. Gstat was not initially
> written for teaching purposes, but for research purposes, emphasizing
> flexibility, scalability and portability. It can deal with a large number
> of practical issues in geostatistics, including change of support (block
> kriging), simple/ordinary/universal (co)kriging, fast local neighbourhood
> selection, flexible trend modelling, variables with different sampling
> configurations, and efficient simulation of large spatially correlated
> random fields, indicator kriging and simulation, and (directional)
> variogram and cross variogram modelling. The S formula/models interface
> is used to define multivariable geostatistical models.
> 
> The source and windows package for R are available from CRAN. The page on
> http://www.gstat.org/s.html has links to R and S-Plus (6.x) source code,
> as well as examples, graphs, and a longer list of features. (A binary
> Win32 S-Plus library is planned later this year.)
> 
> On http://www.ci.tuwien.ac.at/Conferences/DSC-2003/ you can find a
> draft paper further describing the package.
> 
> Any feedback is appreciated.
> --
> Edzer
> 

-- 
Nick Hamm
Department of Geography
Southampton University
Southampton SO17 1BJ
United Kingdom

Office: ++ 44 23 8059 2013 (internal x22013)
Mobile: ++ 44 7887 578 442
Fax:    ++ 44 23 8059 3295
Email:  [EMAIL PROTECTED]