ABSTRACT
Permeability quantifies the efficiency of rocks in transporting fluids; its accurate estimation is crucial for formation evaluation and fluid production forecasts. In shaly sandstones, permeability depends on pore structure and shale or clay content, which may be laminated and/or dispersed. Laminated shale introduces flow anisotropy, whereas dispersed clays constrict pore space, thus reducing porosity and increasing fluid friction. Existing models for permeability estimation typically focus on either laminated shale or dispersed clay, but they are limited in their reliability when both are present simultaneously. We introduce a new permeability estimation model designed specifically for shaly sandstones, accounting for the simultaneous presence of thin shale laminations and dispersed clay. It incorporates the volumetric concentrations of shale lamination and grain-coating clay and accounts for their respective permeability endpoints in conjunction with clean-sandstone permeability to improve the estimation of in situ rock permeability from well logs. Furthermore, we develop an advanced approach for implementing nuclear magnetic resonance (NMR) permeability models, particularly in the presence of thinly laminated shale and immiscible fluids. By integrating our NMR interpretation method with the new permeability model, we significantly improve the accuracy of the estimated formation permeability, hence obtaining a better match between rock core measurements and estimated permeability values. We validate and benchmark the permeability calculated with our model and the newly developed NMR interpretation method against core measurements from a published data set and two field cases of heterogeneous sandstones. In both field cases, we observe a 50% reduction in the deviation of permeability results from core measurements, indicating a significant improvement in the estimation of flow properties despite the associated rock complexity.