MACHINE LEARNING AND THE CONSTRUCTION OF A SEISMIC ATTRIBUTE-SEISMIC FACIES ANALYSIS DATA BASE

Cassel, Murphy

MACHINE LEARNING AND THE CONSTRUCTION OF A SEISMIC ATTRIBUTE-SEISMIC FACIES ANALYSIS DATA BASE

dc.contributor.advisor	Marfurt, Kurt
dc.contributor.author	Cassel, Murphy
dc.contributor.committeeMember	Pigott, John
dc.contributor.committeeMember	Soreghan, Gerilyn
dc.date.accessioned	2018-09-07T19:47:08Z
dc.date.available	2018-09-07T19:47:08Z
dc.date.issued	2018-08-02
dc.date.manuscript	2018-06-14
dc.description.abstract	Currently, seismic facies and structural analysis requires a significant amount of time and effort by skilled interpreters. With the advances made by companies such as Amazon and Google with AI (artificial intelligence) and machine learning, many geoscientists (and perhaps more so, many geoscience managers) have identified the application of such technologies to the seismic interpretation workflow. Advancements of such technologies, such as machine learning based interpretation like self-organizing maps (SOM), principle component analysis (PCA) and independent component analysis (ICA), will both accelerate and quantify the seismic interpretation process. Seismic attributes highlight subtle features in the seismic data that help identify architectural elements that can be used to further define the environment of deposition. Likewise, seismic attributes delineate subtle faults, folds, and flexures that better define the history of tectonic deformation. However, the understanding of “which attribute best illuminates which feature” requires either considerable experience or a tedious search process over years for published analogues. The objective of this thesis is to identify the seismic facies of interest through a prototype a web-based seismic attribute-seismic facies analysis database that can be used not only as a guide for human interpreters, but also to select attributes for machine learning. I propose a rule-based decision tree application that suggests which attributes are good candidates for machine learning applications. There are many seismic facies. This thesis illustrates the objectives and a prototype web application using only two seismic facies: marine mass transport deposits and karst collapses. After initial validation, this product can then be improved and expanded upon by a larger user community to provide an interactive attribute selection platform for interpreters at large.	en_US
dc.identifier.uri	https://hdl.handle.net/11244/301652
dc.language	en_US	en_US
dc.subject	Seismic Facies	en_US
dc.subject	Seismic Interpretation	en_US
dc.subject	Seismic Attribute Application	en_US
dc.subject	Karst	en_US
dc.thesis.degree	Master of Science	en_US
dc.title	MACHINE LEARNING AND THE CONSTRUCTION OF A SEISMIC ATTRIBUTE-SEISMIC FACIES ANALYSIS DATA BASE	en_US
ou.group	Mewbourne College of Earth and Energy::Conoco Phillips School of Geology and Geophysics	en_US