About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

AAPG Bulletin

Abstract

AAPG Bulletin, V. 106, No. 11 (November 2022), P. 2203-2223.

Copyright ©2022. The American Association of Petroleum Geologists. All rights reserved.

DOI: 10.1306/12162121035

Unconventional reservoir characterization by seismic inversion and machine learning of the Bakken Formation

Jackson R. Tomski,1 Previous HitMrinalNext Hit K. Previous HitSenTop,2 Thomas E. Hess,3 and Michael J. Pyrcz4

1Department of Geological Sciences, Jackson School of Geosciences, The University of Texas at Austin, Austin, Texas; [email protected]
2Department of Geological Sciences, Jackson School of Geosciences, The University of Texas at Austin, Austin, Texas; [email protected]
3Department of Geological Sciences, Jackson School of Geosciences, The University of Texas at Austin, Austin, Texas; [email protected]
4Department of Geological Sciences, Jackson School of Geosciences, The University of Texas at Austin, Austin, Texas; Hildebrand Department of Petroleum Engineering and Geosystems Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, Texas; [email protected]

ABSTRACT

Characterization of the critical features of the unconventional Bakken Formation reservoir, specifically within northwestern North Dakota, using three-dimensional (3-D) seismic and well-log data are essential for resource management. We propose a new workflow for the estimation of the features, total porosity, and total organic carbon (TOC) using the output from prestack seismic inversion. Seismic inversion, when integrated with well-log data, can be used to predict these features by first establishing a relationship between the two, and then predicting petrophysical values away from the well-log data. The prediction problem is solved by employing a Bayesian neural network model that uses Markov chain Monte Carlo via Langevin dynamics to sample from the probability distribution and to estimate uncertainty in the desired reservoir features. This method establishes a correlation between estimated P-impedance from seismic inversion and total porosity from well data. Cluster analysis between elastic and petrophysical features is performed to understand the behavior of the reservoir and support the identification of a paleocurrent given the onsite of the Bakken Formation with the initial low surge of P-impedance values. This procedure ultimately allows for the prediction of total porosity and TOC away from the wells at all locations over an inverted 3-D volume, aiding in the assessment of the risk associated with drilling a well. Thus, by integrating our proposed workflow, a better understanding of the features useful for reservoir characterization is possible given a degree of uncertainty, thereby improving oil and gas exploration and risk assessment.

Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Protected Document: $10
Internal PDF Document: $14
Open PDF Document: $24

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].