Electrical Identification of Fractures in Oil and Gas Reservoirs: Rock-Physics Foundations, Resistivity Logging, Borehole Imaging, and Intelligent Interpretation

Abstract

Natural fractures may dominate fluid transport in tight sandstones, carbonates, shales, volcanic rocks, and basement reservoirs despite contributing little bulk porosity. Their electrical signatures are nevertheless non-unique because lithology, fluid salinity, saturation, mud invasion, borehole geometry, and tool sensitivity can reproduce or obscure fracture-related anomalies. This narrative critical review, informed by structured literature searches, links fracture-scale conduction to the responses of conventional resistivity, dual laterolog, multicomponent induction, and electrical borehole imaging tools. The synthesis distinguishes four levels of inference: fracture detection, geometric characterization, potential conductivity, and dynamically confirmed effectiveness. Conductive image traces and deep-shallow resistivity separation can support fracture detection under favorable contrast, whereas position, orientation, and apparent electrical aperture remain tool- and model-dependent estimates. None of these electrical measurements alone establishes reservoir-scale connectivity or permeability. We therefore organize interpretation as a four-stage architecture that conditions the measurements, detects and classifies candidate features, inverts electrical attributes with uncertainty, and tests the resulting hypotheses against core, pressure, mud-loss, sonic, nuclear magnetic resonance, production, or well-test evidence. Automated interpretation may improve consistency, but its field value depends on well-level or field-level external validation and explicit uncertainty calibration. The framework defines which conclusions are directly observed, which are conditional inversions, and which require independent dynamic evidence.