Characterization of Induced Fractures in Carbonate Formation Using Multi-Frequency Dielectric Technique

Abstract

Abstract The multi-frequency dielectric tool is commonly used to estimate water-filled porosity, water salinity, and rock texture. These are estimated from interpretation (petrophysical) models that analyze the dielectric and conductivity dispersions. Previous studies used dielectric permittivity simulators to characterize rock fractures from digital rock models. However, there is no experimental work has been done to characterize fractures from multi-frequency dielectric measurement and validate the simulation results. In this study, dielectric measurements (coaxial probe) were utilized to characterize the geometry of the induced fractures in calcite samples. The dielectric measurements of the samples were acquired before and after inducing the fractures. These fractures were drilled with different diameters and lengths. Also, fractures geometries and locations with respect to a measurement reference are investigated. Afterwards, the dielectric dispersion data were analyzed and correlated with fractures geometry to derive a mathematical model for fractures geometry (aspect ratio) prediction. In the results, after controlling potential factors that affect dielectric dispersion, it is observed that the location of the induced fracture affects the dielectric dispersion. This may hinder the dielectric sensitivity to fractures with different diameters. Although dielectric measurements are more sensitive to the change in fracture length, they could be limited by the depth of the investigation. Based on these observations, several correlations are presented for the fracture aspect ratio of the studied samples. These correlations use the slope and average values of the dielectric and conductivity dispersions at a specific frequency range. This work experimentally demonstrates the sensitivity of the dielectric and conductivity dispersions to fracture geometry. Digital rock models of the studied samples could be developed. Thus, the results of this work may be utilized to calibrate and validate dielectric permittivity simulators to match the actual measurements. Consequently, the need for laboratory measurements could be reduced.