Ice content estimation in the frozen subsurface with an innovative geophysical method: high-frequency induced polarization

The degradation of permafrost due to global warming has pronounced adverse effects, including damage to the infrastructure and climate feedback mechanisms through the release of CO₂. Numerical simulations are often used to predict the speed at which frozen ground is thawing. One essential parameter for these simulations is the ice content, due to its increased heat conductivity and nonlinear behavior during phase changes. Despite its importance, ice content is normally only recorded sporadically as it is difficult to collect measurements at high spatial resolution.

Geophysical methods, which investigate the physical properties of the subsurface, have the potential to estimate the spatial distribution of ice content. Geoelectric measurements are sensitive to the existence of ice since the electrical resistivity of ice is orders of magnitudes greater than that of unfrozen water. However, a quantitative estimate of ice content from resistivity alone is difficult because large resistivities may also be caused by low porosities. Therefore, DC resistivity is sometimes combined with seismic methods to reduce ambiguity.

The high-frequency induced polarization (HFIP) method is capable of measuring ice content as a stand-alone technique. HFIP measures the complex electrical resistivity over a broad frequency range, typically up to 200 kHz. In this frequency range the data is sensitive to an additional property, that is, the dielectric permittivity which represents the material’s capability to be polarized by an electric field. The permittivity of ice exhibits a unique behavior in this frequency range, and therefore HFIP may be used to estimate ice content at the field scale. In order to convert this concept into to a practical method, several challenges must be considered. These include the removal of undesired electromagnetic coupling, efficient data acquisition, the inversion of the raw data to obtain useful images of the subsurface, and the conversion of the frequency dependent resistivity into ice content.

Here, we discuss the progress that has been made in recent years to overcome some of these challenges. Data acquisition relies on the Chameleon II equipment that was designed specifically for HFIP measurements. The Chameleon II is one of the few measuring devices that can perform HFIP measurements on field scale and is able to minimize the undesired coupling between electrical components. Subsurface images are obtained through a 2-D inversion of all frequencies separately, followed by the calculation of ice content using a 2-component model where the electrical properties of the ice fraction and the non-ice fraction are considered. We demonstrate the feasibility of the method using recent case histories from alpine and subarctic permafrost areas. We also show by comparison with independent estimates obtained from drill cores that the obtained estimates are sufficiently accurate. We suggest that this method is ready for practical application and may contribute to the understanding of permafrost degradation processes and to the prediction of the future development of permafrost regions under global warming conditions.