Stratigraphic Intelligence for Hydrocarbon Discovery: Integrating Basin Architecture, Correlation Uncertainty and Predictive Petroleum Systems in Frontier Basins

Salima Cherkeshova, Nurlygul Kylyshbayeva, Gulimshat Shakirova, Kossarbay Kozhakhmet, Samal Janabilova, P. S. Sundetova, Timoth Mkilima

2026

Closed access Metadata (CSL-JSON)
DOI 10.1111/jpg.70144
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

ABSTRACT Frontier hydrocarbon exploration combines sparse subsurface control with uncertainty in stratigraphic correlation, facies continuity, petroleum‐system timing and prospect economics. This structured critical review synthesizes 125 unique publications across platform, rift, intracratonic and other underexplored basin settings spanning Paleozoic to Cenozoic successions. Stratigraphic Intelligence Systems (SIS) is developed here as an integration architecture rather than as a claim that its individual components are new. The framework connects six established or emerging analytical layers: stratigraphic architecture, facies and reservoir‐property prediction, petroleum‐systems modelling, correlation‐uncertainty quantification, near‐surface historical‐geospatial constraints and AI‐assisted data integration. The proposed operationalization represents the subsurface as an ensemble of weighted geological realizations, updates realization weights as new evidence becomes available and propagates the resulting uncertainty into prospect‐level geological chance of discovery and expected‐value calculations. Historical‐geospatial records are restricted to complementary constraints on recent landscape, paleodrainage, uplift or subsidence and source‐to‐sink evolution; they are not treated as direct evidence of Mesozoic basin architecture. AI performance is evaluated within the context of each dataset, label structure and validation design rather than by direct comparison of unrelated accuracy or F 1 values. SIS is therefore positioned as a testable petroleum‐exploration workflow whose value depends on calibrated probabilities, explicit interfaces among model layers and prospective validation against blind drilling outcomes.