About This Item
This article has been peer-reviewed and accepted for publication in a future issue of the AAPG Bulletin. This abstract and associated PDF document are based on the authors' accepted "as is" manuscript.
Editorial Policy for Ahead of Print
Cite This Item
Display Citation
Share This Item
Visit Publisher's Website 
Ahead of Print Abstract
DOI:10.1306/06112422076
Deterministic vs. unsupervised machine learning approach for facies modelling within the Late Devonian Duvernay Formation, Western Canada Sedimentary Basin, Alberta
Elisabeth G. Rau1 , Stacy C. Atchley2 , 
David
W. Yeates3 , Anna M. Thorson2 , and Katherine H. Breen45
1 Matador Resources Company, 5400 Lyndon B Johnson Fwy Suite 1500, Dallas, Texas 75240
2 Baylor University, Department of Geosciences, One Bear Place #97354, Waco, Texas
3 ExxonMobil, 22777 Springwoods Village Parkway, Spring, Tx 77389
4 NASA, Goddard Space Flight Center, Greenbelt, Maryland
5 Morgan State University, Baltimore, Maryland
Ahead of Print Abstract
Facies interpretation from wireline logs has traditionally been performed through comparison of core-observed facies distributions and associated log response from which log-based, deterministic algorithms are developed for facies prediction in wells lacking core control. In contrast, stochastic unsupervised learning analyzes and automatically clusters recurring well log data associations without calibration to core observations. From petrophysical and core description data collected from the Late Duvernay Formation of Alberta, Canada, this study investigates whether unsupervised machine-learning detects lithologic attributes at higher resolution than the traditional deterministic approach. The unsupervised machine-learning methodology non-negative matrix factorization with k-means clustering (NMFk) is applied to identify recurring petrophysical groups independent of core observations. The NMFk groups are compared to four depositional facies associations independently predicted through deterministic assignment of well log cutoffs established by comparing log response to core-observed facies association distributions. Depositional facies associations include the undifferentiated Ireton Shale that overlies the Duvernay, and the open to transitional basin, restricted basin and allochthonous basinal carbonates of the Duvernay. Four NMFk groups are identified: three groups coincide with varying shale lithologies and one group with carbonate lithologies. All groups are similar to the depositional facies associations recognized by the deterministic approach. In addition, NMFk partitions the deterministic restricted basin facies association into subdivisions that are likely distinguished by unique petrophysical responses related to variations in organic richness and associated matrix pore fluid.
Pay-Per-View Purchase Options
The article is available through a document delivery service. Explain these Purchase Options.
|
Watermarked PDF Document: $14 |
Watermarked Document –
A Watermarked Document is branded with the name of the original licensed customer to discourage unauthorized users from sharing the document outside the user's organization. The PDF is no longer restricted to one machine, but can be circulated to others in the same company or department. A Watermarked Document also can be printed for hard copy distribution internally but is not authorized for outside distribution nor posting on the internet. Users will not be able to cut-and-paste text or images from one document to another.
|
| Open PDF Document: $24 |
Open Document –
An Open Document is a fully functional PDF that can be circulated (a digital copy or hard-copy printed documents) outside the purchasing organization. Purchase of an Open Document does NOT constitute license for republication in any form, nor does it allow web posting without prior written permission from AAPG/Datapages
([email protected]).
|
AAPG Member?
Please login with your Member username and password.
Members of AAPG receive access to the full AAPG Bulletin Archives as part of their membership. For more information, contact the AAPG Membership Department at [email protected].
