Premium
On Permeability Prediction From Complex Conductivity Measurements Using Polarization Magnitude and Relaxation Time
Author(s) -
Robinson Judith,
Slater Lee,
Weller Andreas,
Keating Kristina,
Robinson Tonian,
Rose Carla,
Parker Beth
Publication year - 2018
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 - 10.1002/2017wr022034
Subject(s) - permeability (electromagnetism) , exponent , induced polarization , conductivity , electrical resistivity and conductivity , soil science , statistical physics , mineralogy , geology , materials science , mathematics , physics , chemistry , quantum mechanics , membrane , philosophy , biochemistry , linguistics
Geophysical length scales determined from complex conductivity (CC) measurements can be used to estimate permeability k when the electrical formation factor F is known. Two geophysical length scales have been proposed: (1) the specific polarizabilityc pnormalized by the imaginary conductivityσ ″and (2) the time constant τ multiplied by a diffusion coefficientD + . The parametersc pandD +account for the control of fluid chemistry and/or varying minerology on the geophysical length scale. We evaluated the predictive capability of two CC permeability models: (1) an empirical formulation based onσ ″or normalized chargeabilitym nand (2) a mechanistic formulation based on τ . The performance of the CC models was evaluated against measured k ; and further compared against that of well‐established k estimation equations that use geometric length scales. Both CC models predict permeability within one order of magnitude for a database of 58 sandstone samples, with the exception of samples characterized by high pore volume normalized surface areaS p o r . Variations inc pandD +likely contribute to the poor model performance for the highS p o rsamples, which contain significant dolomite. Two observations favor the implementation of theσ ″ ‐based model over the τ ‐based model for field‐scale k estimation: (1) a limited range of variation inc prelative toD +and (2)σ ″field measurements are less time consuming to acquire relative to τ . The need for a reliable field‐estimate of F limits application of either model, in particular theσ ″model due to a high power law exponent associated with F .