## RE: [ai-geostats] a questions of using GSTAT to do Gaussian Condi tional Simulation!

Expand Messages
• ... Hi, Building a stable variogram is like magic ! You have to be a sorcerer, a good one, to deal with everything that matters and make it through the evil
Message 1 of 2 , Jul 8 2:48 AM
On Thu, 2004-07-08 at 08:36, Shazly, Salah DSC92 wrote:
> Dear All,
>
> Are there any rules help building stable variograms, like number of lags,
> range ..etc? And what is the minimum number of control points needed to
> build a variogram?
>
> I am working on a subsurface oil field reservoir.
>
> Regards,
>

Hi,

Building a "stable" variogram is like magic ! You have to be a sorcerer,
a good one, to deal with everything that matters and make it through the
"evil forest of variogram modelling".

In this list there are some of the best sorcerers you can get, so they
may give you a good answer.

There's a very good reference, in my opinion, about geostatistics that
gives you a lot of information on the insights of variogram modelling.

Applied Geostatistics ()
Edward H. Isaaks, R.Mohan Srivastava

It's a "must have" if you want to understand geostatistics and don't
want to deal with all the mathematics. If you do that then check, p.e.

Statistics for Spatial Data (1993)
N. Cressie

You may try to model covariance without using the variogram, with MLE
methods. Take a look at geoR, a package for R.

http://www.est.ufpr.br/geoR/

There's also a paper about using MLS methods directly from the variogram
cloud, check

Muller, W.G. (1999), Least-squares fitting from the variogram cloud ,
Statistics & Probability Letters, 43, 93-98.

Best regards

EJ

--
Ernesto Jardim <ernesto@...>
Biólogo Marinho/Marine Biologist
IPIMAR - Instituto Nacional de Investigação Agrária e das Pescas
IPIMAR - National Research Institute for Agriculture and Fisheries
Av. Brasilia, 1400-006
Lisboa, Portugal
Tel: +351 213 027 000
Fax: +351 213 015 948
http://ernesto.freezope.org
Your message has been successfully submitted and would be delivered to recipients shortly.