Abstract Mudrocks are highly heterogeneous in a range of physical and chemical properties, including: porosity and permeability, fissility, colour, particle composition, size, orientation, carbon loading, degree of compaction, and diagenetic overprint. It is therefore important that the maximum information be extracted as efficiently and completely as possible. This can be accomplished through high-resolution analysis of polished thin sections by scanning electron microscopy (SEM), with the collection of large-area images and X-ray elemental map montages, and the application of targeted particle analysis. A workflow model, based on these techniques, for the digitization of mudrocks is presented herein. A range of the data that can be collected and the variety of analyses that can be achieved are also illustrated. Data collection is discussed in terms of inherent problems with acquisition, storage, transfer and manipulation, which can be time-consuming and non-trivial. Similar information and resolutions can be achieved through other techniques, such as QEMSCAN and infra-red (IR)/Raman spectroscopic mapping. These can be seen as complementary to the workflow described herein.

Abstract The relationship between pore structures was examined using a combination of normalized topographical and topological measurements in two qualitatively different pore systems: organic-hosted porosity, common in unconventional shale reservoirs; and intergranular porosity, common in conventional siliciclastic reservoirs. The organic-hosted pore network was found to be less well connected than the intergranular pore network, with volume-weighted coordination numbers of 1.16 and 8.14 for organic-hosted and intergranular pore systems, respectively. This disparity in coordination number was explained by differences in the pore shapes that are caused by variations in the geological processes associated with the generation of the pore network. Measurements of pore shape showed that the pores in the organic-hosted network were both significantly more spherical and had a more positive curvature distribution than the pores present within the intergranular network. The impact of such changes in pore shape on pore-network connectivity was examined by creating a suite of synthetic pore geometries using both erosion/dilation of the existing network and image-guided object-based methods. Coordination number, Euler characteristic and aggregate porosity analyses performed on these synthetic networks showed that organic-type pore networks become connected at much higher aggregate porosities (35–50%) than intergranular-type pore networks (5–10%).