Understanding the low-frequency electrical properties of ice–water interfaces from laboratory and field experiments

Quantifying ice and water in the subsurface is key to advancing our understanding of permafrost-related processes and to supporting hydrogeological management in cold regions. Electrical resistivity methods are well established as non-invasive tools for discriminating between frozen and unfrozen ground due to the high resistivity of frozen materials. However, solid rocks are also highly resistive, which makes a quantitative estimation of ice content based on resistivity alone challenging. The spectral induced polarization (SIP) method, which allows to measure the conductivity and polarization of rocks and soils, has been proposed as a complementary method for the investigation of frozen ground as ice exhibits a characteristic polarization response in the kHz frequency range due to protonic defects in its crystal lattice. In practice, however, collecting SIP data in such a frequency range is challenging, particularly in alpine environments, due to the logistical constraints and strong capacitive coupling effects related to the high contact resistance between electrodes and the ground. Recently, polarization effects at lower frequencies (< 100 Hz) associated with ice–water interfaces have been reported at the field scale and in laboratory experiments. To quantitatively assess these effects, we present results from experiments conducted at two different scales. First, we show laboratory experiments investigating the SIP response of blank ice features in solid rocks with varying ice volumes and temperatures ranging from +5 °C to −10 °C. Second, we present field SIP data collected during summer 2025 using borehole electrodes in the active rock glacier Muragl (Grisons, Swiss Alps). In the laboratory experiments, holes were drilled into rock samples and SIP measurements were performed after filling the holes with air, water, or ice, allowing a direct comparison of the polarization response for the different pore fillings. We identify a strong polarization effect below 100 Hz for ice-filled holes, while the response remains low when the holes are filled with water or air. Increasing the hole size—and thus the ice volume and the ice–water–rock interfacial area—results in an increase in the polarization strength and a shift of the maximum polarization toward lower frequencies. The SIP field data were acquired using borehole tomography, as borehole electrodes are positioned closer to the ground ice than surface electrodes, enabling a more direct comparison between field and laboratory observations. The results reveal an anomaly characterized by low electrical conductivity and increased polarization starting at approximately 1 Hz in a depth where drillings in summer 2024 revealed relatively high ice content. Overall, our results confirm the presence of a low-frequency polarization effect at ice–water interfaces and demonstrate its potential to image ground ice in alpine permafrost and seasonally frozen soils.