Hi Marius,

There are different ways to deal with a trend, which
depends whether it is considered as global (i.e.
over the entire area) or local (i.e. within the
search window used for kriging). Universal kriging
is a way to deal with local trends, that is the mean
is not assumed to be constant within the search window
but modeled as a linear combination of functions of
spatial coordinates.
For global trends, you can either (1) ignore it if the
search window is not too large with respect to the
extent of the trend and use ordinary kriging,
or (2) model it as a function of spatial coordinates
and perform kriging on the residuals, or (3) use
UK with a very large search window (could not be
implemented if the data density is too high).

The public-domain code ik3d from Gslib would allow you
to implement the aforementioned options.

Regards,

Pierre Goovaerts
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Dr. Pierre Goovaerts
Consultant in (Geo)statistics
and Senior Chief Scientist with Biomedware Inc.
710 Ridgemont Lane
Ann Arbor, Michigan, 48103-1535, U.S.A.

E-mail:  [EMAIL PROTECTED]
Phone:   (734) 668-9900
Fax:     (734) 668-7788
http://alumni.engin.umich.edu/~goovaert/

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

On Tue, 17 Dec 2002, Marius Gilbert wrote:

> Dear colleagues,
>
> I have a set of presence/absence spatial data (1,0) representing the
> locations of site infested/uninfested trees by a tree pest. The
> distribution of infested sites (1) presents a strong trend toward the
> south (infested sites are much more abundant in the south). All
> semivariance estimators (variogram, std variogram, correlogram) are thus
> biased by this trend. I would like to estimate the infestation probability
> using kriging, but wonder how I should adress this. I have considered the
> following options:
>
> a) Indicator kriging: Can IK account for spatial trend ? If so, how  ?
> (any indication regarding software package would be appreciated)
>
> b) Universal kriging of binary data: would this be a valid way to estimate
> infestation probability ? If not, what would be the caveats of using such
> approach ?
>
> Any help would be greatly appreciated and will be summarized to the list.
>
> Have a good day,
>
> Marius Gilbert
>
>
>
>
> --------------------------------------------------
> Dr. Marius Gilbert
> Collaborateur Scientifique FNRS
> Laboratoire de biologie animale et cellulaire
> Universite Libre de Bruxelles CP 160/12
> 50, av F.D. Roosevelt  1050, Bruxelles BELGIUM
> http://lubies.ulb.ac.be
> ---------------------------------------------------
>
>
>
> --
> * 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
>


--
* 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

Reply via email to