## Re: AI-GEOSTATS:

Expand Messages
• Hello Dean The best practical method for this, is plot the variable versus the coordinates. It is like a cross section but with all cross section superimpose.
Message 1 of 34 , Sep 11, 2003
Hello Dean

The best practical method for this, is plot the variable versus
the coordinates.

It is like a cross section but with all cross section superimpose.

This works for 2D or 3D. You make Vr x Coord X or Vr x Coord Y, and so one.

By this way you will see many aspects of your data. Perhaps a global
non-stationarity or only a local non-stationarity.

All of the best

Armando

Dean Monroe wrote:

> Group:
>
> How is the best to graphically show non-stationarity in spatial
> data with say one variable of interest and two dimensions? In time
> series, one can plot realizations over time, what is the spatial analog?
> Second, what may a person do to correct non-stationarity? Again,
> in time series a common practice is to use first, second, etc differences.
>
> Best,
> Dean Monroe
> Oklahoma State University
> Environmental Sciences

[Non-text portions of this message have been removed]
• Thank you for your help What make me think in this problem is this (having to perform groundwater flow modeling): If for example I have two lithological
Message 34 of 34 , Nov 21, 2003

What make me think in this problem is this (having to perform groundwater
flow modeling):
If for example I have two lithological categories with very different
spatial structures ( different anisotropy, different variogram ecc..) but
with the same Hydrogeologic parameters, does grouping these two lithologic
categories together make sense?(for me not too much!)

But, and I'm sorry for my English, what I was asking in my mail was not in
which way I can find the "right" categories. But in which way, once I have
my categories along the boreholes, I can geometrically discretize the
boreholes in data to use in simulation or estimation.
For example I divided a borehole of 10m in punctual values spaced every
25cm....but probably it is not the best approach.

Sincerely
Sebastiano Trevisani

---------------------------------------------------------------------------------------------------
past mails (in geological order!!!!)

Sebastiano

The way we do it is to 'composite' samples with the
same lithology, to a reasonably consistent length.
Then we do semi-variogram analysis either on each
lithology separately or at least not pairing samples
between lithologies. This is an excellent way of
finding out whether your lithologies have different
continuity of values or not. Sometimes you can then
group similar units together.

Depending on how that stage goes, you can then krige
within lithologies or in proper 3d.

Isobel
http://geoecosse.bizland.com/BYOGeostats.htm

________________________________________________________________________
Want to chat instantly with your online friends? Get the FREE Yahoo!
Messenger http://mail.messenger.yahoo.co

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

Sebastiano,

This is a much debated, classic problem reservoir engineers have in the oil
industry. To deal with it, I would recommend you first focus on what you
want your model to do for you. If you plan to build a flow simulation model,
you need to decide how the different rock types affect flow and how they can
be grouped to capture different depletion schemes. Rock types generally can
be grouped together if they have similar properties like porosity,
permeability, and residual saturations. Another way to identify flow units
is to plot (by well) cum well permeability vs. depth or cum well
permeability vs. cum porosity (also called a Lorenz plot), where changes in
the slope of the plot identify breaks between flow units. If these breaks
can be correlated across the reservoir, they make nice boundaries for
geostatistical data points.

Regards,

Paul Taylor
Sr. Reservoir Engineer
BP

----- Original Message -----
From: "sebastiano trevisani" <sebastiano.trevisani@...>
To: <ai-geostats@...>
Sent: Thursday, November 20, 2003 3:21 PM
Subject: AI-GEOSTATS:

---------------------------------------------------------------------------------------------------------------------
> Hi list members!!!
> Another question for you!
>
> I have to perform geostatistical geological modeling using some boreholes.
> The litology in these boreholes is described according to geotechnical
> classification. My question essentially is this: which is the best way to
> transform boreholes in data to use in geostatical modeling? The
litological
> vertical variability is very high, for example I can have layers
> with thickness of only 10cm. If I discretize the boreholes with a coarse
> grid (say a step of 50cm) I lose these details (that in the case of a clay
> layer could be very important from the hydraulic point of view) and if I
> use a too dense grid I have too many data. For example I tried to
> discretize borehole data with a grid, along the borehole, of 10 cm,
> obtaining about 9.000 data. I'm wondering if it could be an idea try to
> discretize in an "intelligent way", for example trying to adapt the step
of
> the grid in relation to the thickness of the level (and why not to the
> category?) encountered. Or if there is another approach...
> I have read some books in geostatistical reservoir modeling...but they
> don't deal with this problem.
> If some has some idea where to look thank you for your help!
>
> Sebastiano Trevisani
>
>
>
> --
> * 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

--
* 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.