AI-GEOSTATS: Questions on GSLIB
- Here under is a previous mail sent to ai-geostats as a word attachment.
Please avoid sending attachments to the list.
Gregoire (Owner of AI-GEOSTATS)
PS: Please reply to "Pablo Vazquez" <mgingenieria@...>
or to the list, not to me.
My name is Pablo V�zquez and I�m working as a Reservoir Engineer for MG&A Oil
and Gas Consulting company from Argentina. Last month we bought a WinGslib
license in order to apply Geostatistic in our work. Now, we are studying some
Geostatistic aspects such as experimental and theoretical variograms, 3-D
kriging estimation techniques and 2-D and 3-D visualizing techniques to build
a static model of an oil reservoir with 38 wells. We are using some log well
data (Gamma Ray, Spontaneous Potential, etc) and we could obtain the
variograms (1 vertical and many horizontal ones), then we fit them with two
different functions (spherical and exponential) and we choose the best one in
a subjective way. Later, we propose a 3-D grid and use the variograms in a 3-D
kriging estimation process. Although we are lack of experience on Geostatistic
we could obtain some valuable information from this analysis. However, we
realize that there are many aspects that we couldn�t study yet.
The purpose of this mail is to ask you some questions about the software and
if it were possible to ask you for some advice. In the following list we show
you some of our doubts:
1. If is it possible to fit a semivariogram in an objective way such as by the
minimization of the quadratic error (the variance or standard deviation)? What
would be your recommendation?
2. If is it possible to obtain an experimental cross-semivariogram with one
variable only? (we ask you this because have seen a work that chose this type
of experimental semivariogram to obtain a Gamma Ray spatial variation)
3. Which are the steps that you recommend in order to do a variogram
validation? Is there another technique than cross-validation?
4. If is it possible to obtain a 3-D kriging output visualization with Gslib
or are there only many 2-D slices? In the last case, do you recommend any
other visualization software?
5. If you only have well log data, what would be the best choice to perform an
oil reservoir characterization: a 3-D kriging estimation technique or a 3-D
Institute of Mineralogy and Petrography
Dept. of Earth Sciences
University of Lausanne
Get free email and a permanent address at http://www.netaddress.com/?N=1
* 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