Premium
Representative Elementary Volumes, Hysteresis, and Heterogeneity in Multiphase Flow From the Pore to Continuum Scale
Author(s) -
Jackson S. J.,
Lin Q.,
Krevor S.
Publication year - 2020
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/2019wr026396
Subject(s) - capillary pressure , imbibition , wetting , relative permeability , multiphase flow , porous medium , hysteresis , saturation (graph theory) , mechanics , capillary action , permeability (electromagnetism) , fluid dynamics , characterisation of pore space in soil , materials science , flow (mathematics) , porosity , geology , statistical physics , thermodynamics , geotechnical engineering , physics , mathematics , chemistry , condensed matter physics , biochemistry , botany , germination , combinatorics , membrane , biology
Representative elementary volumes (REVs) and heterogeneity are key concepts in continuum multiphase flow, yet their manifestation from the pore‐scale and associated impacts with the flow regime are not well understood. We use a multi‐scale experimental and modeling approach to elucidate the role of REVs, hysteresis, and heterogeneity in multiphase flow in two distinct water‐wetting Bentheimer sandstones. Experimental observations during steady‐state drainage and imbibition resolve the pore space at 6 μ m across a field of view up to 12 × 12 × 65 mm with simultaneous measurements of differential pressure during fluid flow. The REV for porosity and capillary pressure is ≈ 2 mm3for both samples, with 5% relative uncertainty. In contrast, due to macroscopic capillary pressure heterogeneities, the REV for saturation varies between 3.4 and 157.5 mm3 , dependent on the fractional flow. Accurate knowledge of this REV uncertainty is critical in assessing continuum scale model validity and predictiveness. We validate the Land trapping model directly by predicting observed non‐wetting fluid connectivity at multiple imbibition states to within the REV uncertainty for homogeneous media. With this, observed hysteresis in the measured relative permeability is entirely removed when considering the connected saturation, validating conceptual hysteresis models based on connected pathway flow. REV scale heterogeneities in capillary pressure are shown to impact flow and trapping characteristics; when built into 3D continuum scale numerical models with hysteresis we are able to predict the variations in trapping and relative permeability observed between the two samples with different heterogeneity structures. The experimental data set herein provides an excellent benchmark for future development of continuum scale modeling frameworks.