No abstract

The Duvernay Formation accumulated as an organic-rich basinal mudrock concurrent with shallow marine platform carbonates of the Leduc and Grosmont formations. Historically classified as a major source rock to conventional hydrocarbon production, the Duvernay evolved into an unconventional shale reservoir across Alberta, Canada much like other source rock intervals worldwide. Distributions of the Duvernay Formation are partitioned into the West and East Shale basins by a narrow, linear Leduc Formation reef complex known as the Rimbey-Meadowbrook trend. Since 2011, development has focused on the West Shale Basin, but thermal maturity trends suggest the potential for expanded shale reservoir development within the southern portion of the East Shale Basin. This study characterizes sedimentologic and stratigraphic controls on Duvernay reservoir potential to identify development “sweet spots” within the East Shale Basin. Duvernay geologic attributes mapped within this study include: oil thermal maturity, thick restricted basin facies association occurrence (at least 5-10m thick), high average TOC values (greater than 2.0 wt.%), and high net carbonate thickness (greater than 40m). The geologic attributes are predictive of production potential within horizontal wells, and the distribution of their co-occurrence suggests the potential for expanded development within the southern portion of the East Shale Basin.

No abstract

Well logs provide insight into stratigraphically compartmentalized rock properties and are a cost-effective alternative to core. The identification of reservoir (and nonreservoir) facies in core, and their calibration to well-log response has traditionally relied on expert domain knowledge and is inherently inconsistent. Such analyses are time-consuming, tedious, error prone, and often biased due to a lack of objectivity. Automated lithologic interpretations from wireline logs appear to be a promising solution for identifying and understanding depositional complexity within a reservoir. Using the Duvernay Formation in the Western Canada Sedimentary Basin as a case study, the authors evaluate the applicability of decision tree-based machine learning (ML) methods in the prediction of core-calibrated facies and/or facies association distributions within wireline logs. The authors use three independent decision tree-based ML models to predict (1) facies (FACM), (2) facies associations (FAM), and (3) reservoir rock (RESM) from wireline logs. Model accuracies are 60.3%, 88.1%, and 88.1% for FACM, FAM, and RESM, respectively, but individual class F1 scores range from 0 to 0.92. The authors attribute discrepancies in individual class performance to interval thickness, sample proportion of training data, and distinguishability of the output class. Classes thicker than 3 m and encompassing at least 16% of the training data set have F1 scores greater than 0.60. The authors attribute exceptions to these general cutoffs to the ability to recognize diagnostic sedimentologic features observed in core. Results from this study help in understanding stratigraphic complexity in the absence of core aiding in subsurface characterization of reservoirs.