On Wednesday 06 June 2007 23:55, Poizot Emmanuel wrote:
Dylan Beaudette a écrit :
> Hi,
>
> I have just posted an example of ordinary kriging using the common
> spearfish data. This is based as close as possible to the most current
> use of GRASS-R bindings. Any criticism is appreciated, as this example
> will be going into other materials... and needs to be as up-to-date as
> possible.
>
> again this is just an example, not a treatise on geostatistics.
>
> http://casoilresource.lawr.ucdavis.edu/drupal/node/438
>
> cheers,
Hi,
Perhaps I would say that to conducte a geostatistic interpolation a best
as possible, it's missing an anysothropy analysis before choosing to fit
a model to your semi-variogram. Looking to your primary results, it
seems that there is variance gradient North-South, so that perhaps the
variogram have to be construct to take it into account if exists.
Regards
------------------------------------------------
Emmanuel Poizot
Cnam/Intechmer
B.P. 324
50103 Cherbourg Cedex
Phone (Direct) : (00 33)(0)233887342
Fax : (00 33)(0)233887339
------------------------------------------------
Thanks for the comments Emmanuel. I agree that there should always be an
investigation into all of the factors which might otherwise invalidate the
intrinsic hypothesis of kriging [1].
I have included a couple extra code snippets and figure into the main material
that I am working on. I will integrate this into the original article, until
then - here is a related page:
http://casoilresource.lawr.ucdavis.edu/drupal/node/442
Cheers,
PS: your feelings about the anisotropy matched mine as well, along with the
attached variogram map
1. Issaks, E.H. & Srivastava, R.M. An Introduction to Applied Geostatistics
Oxford University Press, 1989
--
Dylan Beaudette
Soils and Biogeochemistry Graduate Group
University of California at Davis
530.754.7341
(attachments)
variogram_map.pdf (11 KB)