Rockwall permafrost dynamics evidenced by Automated Electrical Resistivity Tomography at Aiguille du Midi (3842 m a.s.l., French Alps)

Abstract. Permafrost warming significantly affects the stability of rockwalls in high altitude regions. Subsurface monitoring of permafrost is essential to assess the resulting potential geohazards. This study investigates permafrost dynamics at Aiguille du Midi (3842 m a.s.l., French Alps) using an Automated-Electrical Resistivity Tomography (A-ERT) approach, conducted over a period of four years (2020–2023). A total of three geophysical profiles have been installed on three sides of the Aiguille du Midi. An autonomous acquisition system for permanent resistivity monitoring and remote data acquisition is implemented. A time-lapse inversion technique is employed to get time lapse variations of the electrical resistivity of Aiguille du Midi at different time scales. In addition to the field measurements, laboratory measurements of electrical resistivity are conducted on one water-saturated granite sample in both unfrozen and frozen conditions to evaluate the temperature-dependency of resistivity. Temperature information about the thermal state of permafrost is available from three shallow boreholes drilled in 2009, used to validate our interpretation. A-ERT showed significant variations in the active layer thickness across different rock faces, along with a slight decrease in the resistivity of permafrost, indicating its warming over time. Our findings indicate that the temperature dependence of resistivity in field conditions (open system) is slightly less pronounced than in controlled laboratory experiments (closed system). Using a petrophysical model connecting temperature to resistivity, the temperature distribution within the frozen zone can be estimated from the resistivity measurements (during summer and autumn) with an accuracy of ~±1 °C. This research underscores the efficacy of ERT as a promising, non-invasive tool for monitoring permafrost dynamics in Alpine environments. It highlights the need for further methodological refinement to better resolve subsurface properties, potentially using induced polarization data.