- View SourceThe estimation variance is the answer. You can to estimate using Kriging the

standard error of estimation that is the standard deviation/mean or estimate

by kriging. Remember that the standard deviation of kriging can vi see as a

usual standard deviation and it means the statistical rate of possible

variation of the estimate value by kriging: ([Z*-2?K]<=Z* <=[Z*+2?K])

The possible solution of location of the news points is to do a map of the

variance of estimation and select the maximums for locates it. Other

possibility is the algorithm of the fictitious point, it is an iterative

algorithm that locate successively a point in the place of biggest variance

of estimation and repeating the algorithm with the new fictitious point

(remember that estimation variance do not depend of the value of the point,

depend of it location and the variogram)

If you use only 15 location or points, and those fluctuations are not to big

you can use the topography information for build the variogram.

It can be interesting to compare your results with universal erosion

formulae.

King regards

Adrian Martínez Vargas

Ing. Geólogo (profesor)

Instituto Superior Minero Metalúrgico (ISMM)

Las Coloradas, s/n

Reparto Caribe.

Moa, Holguín, Cuba

CP 83329

E. Mail

amvargas@...

> -----Mensaje original-----

--

> De: Veerle Huvenne [mailto:veerle.huvenne@...]

> Enviado el: martes, 12 de febrero de 2002 9:51

> Para: ai-geostats@...; Geert.Moerkerke@...

> Asunto: AI-GEOSTATS: sampling strategy

>

>

>

> Hello Ai-geostats list members,

>

> A collegue of mine has the following question concerning sampling

> strategy :

> Given : a test site at which the effect of

> sedimentation/erosion has to

> be checked regularly (say every year). To start off with, the

> topography

> of the area is known. The first year the elevation is

> measured in a set

> of discrete points (say some 10 to 15 locations). From this one can

> gather already some information as to where there has been

> sedimentation

> and erosion. The following years, measurements are planned at the same

> locations. However, now comes the question :

> is there any rule of thumb, theory, calculation,... which allows to

> decide if more or less samples/measurements are necessary to achieve a

> certain precision in the mapping of erosion/sedimentation, given the

> information which can be derived from the measurements taken in the

> first year?

> If it is needed to plan more sample points, where should they

> be placed?

> In the areas of highest sedimentation/erosion? Or is it better to just

> choose a denser sampling grid?

>

> Has anybody any information on this? I know it's always difficult to

> plan a sampling design because one does not know what one will find in

> the test site, but this time there is some preliminary information

> already due to the measurements made during the first year.

> It seems to

> me this might change the way of looking at the problem. Or not?

>

> Looking forward to your answers, and thanks for your help

>

> Veerle

>

> --

> Veerle Huvenne

> Renard Centre of Marine Geology

> University of Ghent

>

> Krijgslaan 281, S8

> 9000 Gent, Belgium

> +32/9/264.45.84

>

>

>

> --

> * 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

>

* 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