IMPROVED UNDERSTANDING OF EFFECTS OF SOIL SATURATION ON SPECTRAL INDUCED POLARIZATION

Spectral Induced polarization (SIP) is an emerging technology that provides unique information on the physical and chemical properties of the interconnected pores and pore surfaces. While previous studies focused on fluid chemistry and matrix composition, more research into the effects of saturation (θ) and matric potential on SIP is needed. The objective of this research was to determine the effects of unsaturated water content dynamics on SIP parameters. One hypothesis is that specific polarizability, the interfacial polarization normalized by specific surface area, requires consideration of hydrodynamic conditions.

To test this hypothesis, SIP measurements were conducted over a range of moisture contents on a laboratory grade sand, an undisturbed soil consisting of silty sand, and a prepared sample consisting of a mixture of sand and field soil. A sample apparatus was designed that allowed for electrical measurements at varying water contents, while also permitting measurements of in situ water tension and θ. The specific surface area per unit pore volume of the two soils was measured using the nitrogen BET method.

Decreases in θ resulted in a decrease in real conductivity (σ’) due to the removal of conductive pore solution. In addition, a decrease of imaginary conductivity (σ”) was also observed, indicating that specific polarizability is reduced as water is removed from the soil. To evaluate this affect, we compared our data to predictions from two models. The first model is based on the polarization of the Stern layer and Maxwell-Wagner polarization and predicts an increase in amplitude of complex resistivity with decrease in θ. The second model relates mobility differences to the variation in apparent pore sizes and predicts an increase in normalized chargeability followed by a decrease at low θ. In the field soil, our observations are in apparent contrast to published data of unsaturated SIP measurements. However, previous studies applied static hydraulic conditions only during SIP. We propose that the greater decrease in σ” then σ’ during initial drainage, is likely due to dynamic hydraulic conditions in our set up, indicating the importance of experimental conditions for SIP. This data provides a better understanding of how SIP can be used to determine moisture dynamics in laboratory and field investigations.