Experiments on the complex conductivity signatures of pore pressure variations in soils

Understanding how pore pressure variations impact soil stability is crucial for studying geohazards like landslides, as these events alter pore pressure (PP) distribution in the subsurface and often are triggered by sudden changes in PP distribution. However, obtaining in situ PP measurements at large scales is challenging. Geophysical methods provide an indirect means of assessing PP changes in soils. In this laboratory study, we examine the complex conductivity (CC) signatures of soils subjected to increasing pore pressure. A rigid acrylic sample cylinder was equipped with a pair of current electrodes at the top and bottom, along with a pair of potential electrodes on the side. We experimented on different synthetic soil mixtures as well as a natural soil sample collected from central Oklahoma. The synthetic soils were mixtures of silica sand and various clay percentages (0%, 3% and 11% by weight). Montmorillonite and kaolinite clays were used to explore the effects of matrix expansion versus no expansion in this study. All soil samples were fully saturated with a 1 mS/cm potassium chloride solution. We measured the CC response of soil as we increased the pore pressure in 10 kPa increments starting from atmospheric pressure up to 200 kPa. Our results show that the CC method is sensitive to changes in pore pressure values with imaginary conductivity magnitude increasing with increasing pore pressure for the samples containing clay minerals. The pure sand soil sample showed a less pronounced yet similar trend to clayey mixtures. The natural soil sample and samples with montmorillonite showed a direct relationship between imaginary conductivity and PP while real conductivity and PP showed an inverse relationship. In the samples without montmorillonite, we observed no changes in the characteristic relaxation time (τpeak) indicating no pore geometry changes in these samples. However, the samples with montmorillonite showed a direct linear relationship between PP and τpeak. Our findings indicate that the CC measurements are suitable for tracking PP changes in soils; however, further research in the field, and with a wider spectrum of soil types and larger PP increments, under more natural conditions, is needed to validate our results.