Pore structure is one of the main factors defining reservoir properties of rocks. Conventional estimations of such properties using only limited 2D data from thin sections using different approximations are usually inaccurate.
In recent decades different numerical methods were developed to quantify flow and other physical properties on micro-scale. Each method has its own advantages and drawbacks, so utilization of more than one of approaches is reasonable, depending on the study case.
Recent progress in X-ray micro-tomography and some other techniques allow precise determination of three-dimensional structure of rocks, however, a trade-off between resolution and sample size is usually unavoidable. Some amount of porosity (we call it "under-resolution") is usually not visible on X-ray scans or thin-section low resolution images. Unlike permeability, it may play an important role in resistivity, capillary curve shape and other properties.
The main aim of this contribution is to verify petrophysical modeling approach on a collection of sandstone samples with wide range of pore space configurations. At first, our 3D structure obtaining method using X-ray microtomography is justified via detailed laboratory vs. tomography porosity measurements comparison. Next, permeability is determined for all samples using network-model extracted from 3D structure scans. Calculated formation factors and capillary curves in many cases deviated from experimental values, especially for samples with high under-resolution porosity. The influence of invisible porosity is also supported by coordination number correlations with other physical properties for sample within the same group (same reservoir type). To account for under-resolution pores invisible on X-ray scans different approaches can be utilized based on artificial addition of under-resolution porosity into numerical sample: NMR measurements, mercury porosimetry, or stochastic reconstructions from high resolution 2D cuts.