Lithofacies classification in the Marcellus Shale by applying a statistical clustering algorithm to petrophysical and elastic well logs

Abstract

Interpreting shale lithofacies is an important step in identifying productive zones in the Marcellus Shale gas play; the target reservoir can be less than 3 m (10 ft) thick with overlying shales that appear similar on certain petrophysical well logs and in the core. However, these nonreservoir-quality shales contain widely varying organic and mechanical properties. To distinguish between reservoir- and nonreservoir-quality shale facies, this classification method applies a pattern-recognition algorithm, expectation maximization, to a set of commonly available petrophysical and elastic well logs to create 1D shale facies models along the wellbore. The statistical algorithm classifies the petrophysical and elastic well-log data into defined facies using the theory that each well log contains a distribution of Gaussian curves for each facies. The method is applied to 12 wells across Pennsylvania and northern West Virginia, and it is able to discriminate between shale facies based on organic content and brittleness, characteristics which are not always evident in core. To verify the geologic accuracy of the facies models, the results are compared with the core data, well logs, mudlogs, and regional stratigraphic studies when available. In addition, we were able to designate a lithofacies-defined top to the Marcellus Formation in parts of the Appalachian Basin where the gradational contact with the overlying Mahantango Formation is poorly defined on petrophysical logs.

A novel prestack sparse azimuthal AVO inversion

“…We expect that propagating this uncertainty will lead to a skewed estimate of B iso . Finally, assuming stationarity in φ s , we choose the solution for the isotropic gradient that is consistent with the regional analysis of (Sclanser et al, 2016).…”

“…Similarly, applications of A.I. to well-log analysis are already well established in the geophysical literature, (Hall and Hall, 2017) and (Sclanser et al, 2016) are recent examples supervised and unsupervised learning applications. The key feature in the application of A.I.…”

“…Assuming Ruger's model, we derive an estimate of the anisotropic gradient, and fracture orientation, up to a 90 • ambiguity. In a recent study by (Sclanser et al, 2016), the EM-algorithm was applied to well-log data for the purpose of classifying litho-facies. This study identifies four classes of shale, and a carbonate in our area of interest.…”

“…The coefficients of Ruger's model are determined using the method described above. The ambiguity in the sign is resolved by ensuring that the isotropic gradient is consistent with regional data derived from (Sclanser et al, 2016). The model parameters for base of Marcellus are shown in Fig.…”

A Comparative Study of Five Networks for Reservoir Classification Based on Geophysical Logging Signals

“…For examples, Cui et al [6] applied the principal component analysis (PCA) method on multiple logging signals to classify four diagenetic facies pre-defined by samples from a tight sandstone reservoir in the Ordos Basin, Central China. Schlanser et al [7] tested a statistical clustering algorithm with geophysical logs for lithofacies classification in the Marcellus Shale. Guo et al [8] proposed a combination model--three ANNs to predict porosity, permeability and shale content respectively following a neuro-fuzzy inference machine--for pay zones recognition.…”