Multidisciplinary Characterization and Modeling of Mississippian Carbonate and Silica-Rich Reservoirs, Northern Oklahoma

Abstract

Mississippian carbonate and silica-rich deposits of the mid-continent formed on a regionally extensive carbonate ramp and form shoaling-upward lithofacies successions that stack into high-frequency cycles. The cycles form several prograding depositional packages that downlap to the south. Lithologies include tripolitic chert and limestones that vary from mud- to grain-dominated fabrics. Tripolitic chert, the primary reservoir, most likely formed during periods of exposure; however, hydrothermal processes are also a possible cause. While the chert is most common at the top of the Mississippian, other cycles are capped by high-porosity, low resistivity chert. Lithofacies from core are calibrated to well logs using various methods and tied to seismic attributes to predict and map their spatial distribution. Support Vector Machines (SVM), a supervised-learning method in data mining, learns and recognizes the patterns that exist between the core and log datasets. SVM analyzes lithofacies (from core) and logging response and produces a model for lithofacies that is used to predict lithofacies in non-cored wells. The lithofacies logs are used to establish the stratigraphic framework and are integrated with seismic data (attributes) to condition 3-D reservoir models of lithofacies and petrophysical properties. The models are important to show depositional, diagenetic, and structural trends, analyze reservoir connectivity, and relate production to these characteristics.

AAPG Datapages/Search and Discovery Article #90221 © 2015 Mid-Continent Section, Tulsa, Oklahoma, October 4-6, 2015