The recognition of clastic parasequences is hampered by vagueness in their definition and rendered difficult by variability in their origin and temporal and spatial expressions. This makes the automation of parasequence identification in subsurface data particularly challenging. This study presents a supervised machine-learning approach for the recognition of parasequence boundaries (flooding surfaces) in borehole data. The approach enabled the classification of facies logs using a training data set consisting of over 9000 facies units in 169 shallow-water clastic parasequences, from both outcrop and subsurface geologic analogs. Data on the thickness, grain size, and vertical stacking trends of sedimentary facies were collated to produce three alternative training data sets based on vertical windows (facies sequences) spanning a variable number of facies units (two, three, and five). Eight alternative machine-learning algorithms for supervised classification were employed to perform picking of parasequence tops, a task that was treated as a binary classification problem. The training of the models was performed by optimizing their ability to detect positive instances; i.e., portions of facies sequences that were correctly identified as occurring at parasequence tops. Model evaluation against a data set reserved for testing revealed that ensemble machine-learning algorithms were able to pick flooding surfaces that matched those recognized by stratigraphers in up to 39% of the cases. Generally, sensible classifications were made by the best performing models where evident thickness and grain-size signatures of flooding surfaces existed in facies sequences. Because of the aforementioned subjectivity in parasequence definition, model training and performance assessment were not straightforward. Nevertheless, promising aspects of this approach include: (1) its applicability to ensemble machine-learning algorithms; (2) the ability to train classifiers on outcrop analogs; (3) the capacity to compile different training data sets reflecting different schools of thought about parasequence definition and hierarchical significance; and (4) its possible integration with computational methods for classifying lithologies in borehole data and for correlating surfaces across wells.

Stratigraphic geometrical databases are critical tools for predicting extents of reservoirs, heterogeneities, and seals in hydrocarbon, hydrogeological, or carbon capture and sequestration workflows. Any attempt at generating stratigraphic architectural geometrical databases relies on classifications and anchoring of architectures such that self-similar features cannot float across both spatial and temporal hierarchical scales. Any such movement across hierarchical scales invalidates the applicability of the data collated. The parasequence is a classic example of an architectural unit that floats across hierarchical scales. Since no scale is specified in its definition, users have the freedom to apply the term at different scales. The approach detailed herein aims to avoid the issue described above by anchoring the parasequence at the regressive–transgressive, full or partial shelf-transit scale. Once the anchor has been applied, recognition of consistent, hierarchical, self-similar stratigraphic architectures internal to the parasequence results in a consistent method for recognition of parasequences working from the smallest-scale unit upwards via a “bottoms-up” approach. In shoreline depositional systems, there are three key architectural units recognized in vertical profiles within the parasequence: 1) the bed-set or Element (E), 2) the Element Set (ES) which represents the beach-ridge set or mouth-bar set in plan-view, and 3) the Element Complex Set (ECS) which represents the plan-view coastal product of one delta-plain fluvial-channel avulsion. Each of these architectural units and thereafter, the parasequence is defined by a set of vertical thickness patterns and trend rules that can aid in identification of the discrete stratal units in vertical pro-files. Whilst enabling collation of data at appropriate hierarchical levels, the methodology also provides a technique for enhancing chronostratigraphic unit recognition and correlation at high resolution, intra-parasequence scales, and for improved prediction of the distribution and geometries of heterogeneities that influence fluid flow.