About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

Oklahoma City Geological Society

Abstract


The Shale Shaker
Vol. 63 (2013), No. 5. (March/April), Pages 326-339

Characterizing a Mississippian Tripolitic Chert Reservoir Using 3D Unsupervised Multi-attribute Seismic Facies Analysis

Atish Roy, Benjamin L. Dowdell, Kurt J. Marfurt

Abstract

Supervised and unsupervised seismic facies extraction methods are slowly gaining popularity in the hydrocarbon exploration and production workflows. The unsupervised clustering is data driven, is not biased by the interpreter beyond the choice of input data, and brings out the natural clusters present in the data. There are several competing unsupervised clustering techniques, each having advantages and disadvantages.

Our 3D multi-attribute analysis is based on Kohonen self-organizing maps (SOM), which are one of the most commonly used unsupervised classification algorithms. The Mississippi lime resource play of northern Oklahoma and southern Kansas provides a particularly challenging problem. Here, the seismic facies are diagenetic (limestone, non-porous chert, and highly porous tripolitic chert) or structural (fractured chert and limestone) rather than stratigraphic. Skilled interpreters can recognize the different facies although they are not able to accurately define their diagenetic or tectonic history. During the past several years, over 4700 wells have been drilled in the Mississippi lime play. Such dense drilling provides a means of calibrating the computer-generated “natural” clusters that represent the data. Using a 3D seismic survey acquired in Osage County, Oklahoma, we use SOM to represent the Mississippi lime play. The 256 prototype vectors (potential clusters) reduce to only three or four distinct clusters. The predicted clusters correlate well with an independent facies classification using post-stack impedance inversion. This workflow automatically differentiates depositional packages characterized by subtle changes in the stratigraphic column as well as lateral changes in texture.


Pay-Per-View Purchase Options

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

Watermarked PDF Document: $14
Open PDF Document: $24