Hi there

About a month ago I posted this message asking about GSLIB's Gaussian

Simulation program. I had one reply [thanks] but I need write up an

outline/derivation of the algorithm. Neither of Isaaks & Srivastava or

Wackernagel appear to deal with simulations.

** Are there any references that explain the Gaussian Simulation algorithm

with the readability of I&S or W? **

Thanks

My original post was:

>

> Can someone help me out with a question.

>

> My understanding of GSLIB's sgsimm Sequential Gaussian Simulation program

> is that it produces a gridded map of values, each of which comes from a

> Gaussian [Normal] distribution, and as a whole has an experimental variogram

> much more similar to the sample experimental variogram than that of ordinary

> kriging.

>

> Q1: Is this true?

> Q2: if so, how do they do it?

>

> I am guessing that at each point, z_i~~N(mu,sigma). I presume the parameters

> are the kriged value and kriging variance. That seems reasonable, except

> that I thought the kriging variance wasnt a very good measure of estimation

> error??

> Accepting that, how do they impose the constraint on the variogram produced

> by the simulation??

>

> Any help will be gratefully accepted.

>

> I apologize if this is a RTFM-type question.

>

> uoY knahT

>

> Andrew

> --

--

--------------------------------------------------------------------------

Dr Andrew Gill Room:203d

Department of Applied Mathematics Email:

agill@...
Mathematics Building Phone: 08-3035577

The University of Adelaide Fax: 08-3033696

Adelaide, SA, 5005

--------------------------------------------------------------------------

--

*To post a message to the list, send it to

ai-geostats@....

*As a general service to list users, please remember to post a summary

of any useful responses to your questions.