RE: AI-GEOSTATS: Kriging with External Drift

[Michael Dennis]

Title: RE: AI-GEOSTATS: Kriging with External Drift

Thanks for clearing that up for me.  I get confused with the terminology sometimes.  I'll see if I can get a copy of the book you mention to look at to see if it helps me out.

 

You say that KED uses a shape function for the "Drift" data.  How is this shape function computed?  Isn't it just a polynomial trend?  Or is it something more complex than that?  I know the bottom row of the matrix is computed from the "drift" but I'm interested in how that bottom value is derived.

 

On another note : I've run into a couple of references on Bayesian Kriging which I get the impression is similar to KED.  Can anyone tell me what the differences are between Bayesian and KED and why you would choose one over the other?

 

Thanks,

 

Mike

 

 -----Original Message-----
From: WARR Benjamin [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 06, 2001 3:59 AM
To: 'Michael Dennis'
Subject: RE: AI-GEOSTATS: Kriging with External Drift

Michael,

I suggest a book by Hans Wackernagel, An Introduction to Multivariate Geostatistics, Springer Verlag, 2nd Edition 1999.

He describes the technique thoroughly and clearly. Briefly, what you describe is Universal Kriging (UK). KED uses a shape function, provided by exhaustive secondary data to model a deterministic 'drift' or underlying trend, whereas when doing UK you have to model the trend yourself using a polynomial.

Benjamin Warr

Research Associate
Centre for the Management of Environmental Resource(CMER)
INSEAD
Boulevard de Constance,
77305 Fontainebleau Cedex,
France

> -----Original Message-----
> From: Michael Dennis [mailto:[EMAIL PROTECTED]]
> Sent: 06 December 2001 04:13
> To: AI-Geostats Mailing List
> Subject: AI-GEOSTATS: Kriging with External Drift
>
>
> Hello all,
>
> I'm a bit of a rookie with GeoStatistics.  I am interested in
> Kriging with
> External Drift but I am having a hard time finding
> information that tells
> you how it works in laymans terms (without just firing a
> matrix at you and
> leaving you to deduce what it means).
>
> I don't think this is right but I'll give it a shot.  Does it work as
> follows :
>
> 1) Compute a trend for the drift variable
> 2) Remove the trend computed in 1) from the main variable
> 3) Grid the residuals from step 2)
> 4) Add back the trend from step 1)
>
> I don't think this is right but if you can explain to me how
> the Drift is
> actually applied in laymans terms it would be greatly
> appreciated.  Also
> when you do kriging with external drift do you have to model
> a variogram or
> can a reasonable one be computed automatically, if so how
> would you compute
> it?
>
> Also if you have any good sources of information on Kriging
> with external
> drift could you pass them on to me?
>
> Thanks,
>
> Mike
>
>
> --
> * To post a message to the list, send it to [EMAIL PROTECTED]
> * As a general service to the users, please remember to post
> a summary of any useful responses to your questions.
> * To unsubscribe, send an email to [EMAIL PROTECTED] with no
> subject and "unsubscribe ai-geostats" followed by "end" on
> the next line in the message body. DO NOT SEND
> Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org
>