This study demonstrates applications of core and advanced well logs to computing general log-based stochastic multimineral solutions to build detailed 1D shale petrofacies model and integrate with chemostratigraphy to better decipher depositional environments of the Bakken Shale units in the Williston Basin of North Dakota, USA. In particular, relationships between trace element geochemical data and organic matter coupled with well-log-derived crossplots and solutions are explored to understand vertical and areal heterogeneity of the shale members in the Bakken Formation. A methodology based on mineral composition and organic-matter richness derived from well logs and core data is proposed for facies classification in the Bakken mudstone units. The results show that Bakken shale members are heterogeneous, in terms of mineralogy and organic matter, which can be classified as five different petrofacies, reflective of changes in depositional and diagenetic environment. Highly organic-rich shale facies units were deposited in euxinic environment, whereas relatively organic-poor shale units were deposited in anoxic and dysoxic conditions. Statistical analyses suggest that trace element geochemical data can be applied to a significant degree of confidence to compare with log-derived facies model to characterize different shale petrofacies and construe the depositional environment in detail.
Lithofacies or petrofacies classification, assigning a rock type to specific rock samples on the basis of petrography or measured petrophysical properties, is fundamental to subsurface investigations. Clastic and carbonate petrofacies have been studied extensively for depositional and diagenetic environment studies. However, research in black shale petrofacies is relatively rare, most being based on either single well studies or descriptive analysis (Schieber, 1999; Hickey and Henk, 2007; Egenhoff and Fishman, 2013; Bhattacharya et al., 2015). A case study from the Bakken Formation in the Williston Basin in North America has been chosen for this study.