Characterizing landslide dynamics from time-lapse time domain induced polarization and ground-based imaging: a case study of the MontGombert landslide (French, Alps)

Electrical resistivity tomography (ERT) is a common tool to study landslide morphology and dynamics. Landslide dynamics is strongly related to clay content, pore pressure and permeability distribution. The standard direct current (DC) resistivity method often fails at providing information about permeabilities because materials with low resistivities can both exhibit high or low permeability especially according to their clay content and the relative importance of surface conduction compared to electrolytic conduction. Induced polarization (IP) has proven to be a useful complementary tool to assess petrophysical properties of the rocks. Up to now, it has been rarely used in landslide studies and not yet in landslide monitoring. This study aims at demonstrating the usefulness and added value of IP as a complement to traditional DC measurements in the case of landslide characterization and monitoring. We performed 3 time domain IP (TDIP) surveys at three different time steps on an active landslide in the French Alps. A time-lapse inversion of the resistivity and chargeability is realized. The spectral IP parameters are extracted through a Debye decomposition. The results are compared to surface displacements obtained by processing optical images. We demonstrated that the use of TDIP is of crucial importance to estimate the extent of unstable volumes within the landslide. The results are in very good agreement with the data provided by the piezometers and inclinometers. The use of a general law to link IP variables and permeability enables to estimate the spatial distribution of the permeability within the landslide and its evolution through the seasons.