Hi Luke
Yes, as you suggest very often in geology the third dimension (say z) is
more similar to a temporal dimension than to a spatial one.
(Think for example to quaternary sedimentary deposits). So is it so a big
deal to preform a 3d analisys? isn't it better to conduct
a stratified 2D analysis?
For this reason I`m glad if someone has some idea also only for 2D
cross-variogram fitting.
Bye
Sebastiano
At 12.19 01/03/2006, Luke Spadavecchia wrote:
Hi seba,
I have found similar problems. What kind of 3D model are you trying to
fit? If it is a separable covariance model (of the product or product-sum)
variety, a good place to start is the modified GSLIB code that appeared in
the attached article. The code can be downloaded from the computers and
geosciences website. The code will compute the empirical semivariance
surface for you, but there is no fitting routine. I am about to embark on
a major programming task (when I get the time) to fit a variogram surface
to the output of GAMSURF, but in the meantime, I just fit my model in
Excel, or some other spreadsheet. I have attached an example for you:
although it is spatio-temporal, the same model is permissable in 3D space.
Any nested variograms permitted in 2D can be used for the separable
models. This is actually quite appealing, given the nature of the
covariance in the vertical may be different in functional form from the
horizontal covariance, since different processes may underlie the
autocorrelation.
Hope that is helpful
Luke
* By using the ai-geostats mailing list you agree to follow its rules
( see http://www.ai-geostats.org/help_ai-geostats.htm )
* To unsubscribe to ai-geostats, send the following in the subject or in the
body (plain text format) of an email message to [EMAIL PROTECTED]
Signoff ai-geostats
