## AI-GEOSTATS: Re: Urgent help needed: Power-law Semi-variogram Range

Expand Messages
• Thank you for your reply, By the transect method, I mean by taking a straight line through the field of observation. The transect method computes the variogram
Message 1 of 2 , Feb 11, 2004

By the transect method, I mean by taking a straight line through the field of observation. The transect method computes the variogram along this line only, since it is an isotropic process generated from a functional form.

Is there a reliable method for determining the range(correlation length).
I wrote a function in Matlab to determine and plot the variogram. I cross-checked by running the generated data through GSTAT --- it also results in a power law variogram.

Please advise on the best approach to determine the correlation length in this case.

Thanks

Joseph.

----- Original Message -----
From: Gerald Boogaart <boogaart@...>
Date: Wed, 11 Feb 2004 11:36:20 +0100
To: seacode@...
Subject: Re: Urgent help needed: Power-law Semi-variogram Range

> Dear Joseph,
>
> I don't know wether I can help, not knowing what you mean by transect method.
> Anyway others can give better answers later.
>
> As far as I understand the subject there is no range for a power law
> variogram, since it is unbounded. Anyway some software will ask for a range,
> which is that either ignored or used in some softwarespecific way e.g. in the
> way gamma(h)=nugget + sill* (h/range)^p where it is redundent with the sill.
> Just try to find out, and fit visually or reask the question specifing the
> software you are using. Anyway when the software is asking for a sill or
> range in power law variograms the software uses some odd convention.
>
> Best regards,
> Gerald
>
> On Tuesday 10 February 2004 21:04, you wrote:
> > Hi Geostats,
> >
> > Kindly advise me on how to determine the range value from a power-law semi
> > variogram. I am new to geostatistics but I need to analyse sample data from
> > an isotropic process. I am getting a power-law. How can I determine a range
> > value? I am using the transect method.
> >
> >
> > Joseph
>
>

--
___________________________________________________________
Sign-up for Ads Free at Mail.com

--
* To post a message to the list, send it to ai-geostats@...
* 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 majordomo@... 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
• Joseph, not all spatial processes have correlation lengths; some processes show an increase in variance whenever you increase the domain. Think of (1D case)
Message 2 of 2 , Feb 12, 2004
Joseph,

not all spatial processes have correlation lengths; some
processes show an increase in variance whenever you
increase the domain. Think of (1D case) Brownian
motion. Power law variograms result from processes
that have no correlation length; IIRC, Brownian
motion has a linear variogram (power=1) with no sill.
--
Edzer

Joseph Oyeyele wrote:

>
>By the transect method, I mean by taking a straight line through the field of observation. The transect method computes the variogram along this line only, since it is an isotropic process generated from a functional form.
>
>Is there a reliable method for determining the range(correlation length).
>I wrote a function in Matlab to determine and plot the variogram. I cross-checked by running the generated data through GSTAT --- it also results in a power law variogram.
>
>Please advise on the best approach to determine the correlation length in this case.
>
>Thanks
>
>Joseph.
>
>
>

--
* To post a message to the list, send it to ai-geostats@...
* 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 majordomo@... 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
Your message has been successfully submitted and would be delivered to recipients shortly.