Temperature influence on relative seismic wave velocity measurements for landslide monitoring

Abstract

Listen

Translate article

Seismic interferometry is used as an innovant method since a tens of years to monitor landslides (LeBreton et al., 2021; Larose et al., 2022), especially since a sharp decrease of about 7 % in relative seismic velocity have been measured on the Pont-Bourquin clay landslide (Switzerland) a few days before a significative reactivation of the instable slope (Mainsant et al., 2012). However, if the relative variation of seismic wave velocity (dV/V) appears as a good precursory signal before a failure, indicating a  loss hydro-mechanical properties of the subsurface, the relative variation of seismic wave velocity (dV/V) within time generally show reversible variations due to environmental forcings. In the present work, we analyse more than two years of seismic data measured on a large clayey morainic landslide located in Valloire (Savoie, France), where seismic interferometry was coupled with other surface displacement monitoring involving RFID measurements and time-lapse photographies (Laigle et al., 2019). On this site and during the stable considered period, dV/V appears well correlated with air temperature but in an unexpected way, since a decrease of temperature is correlated to a decrease of dV/V, instead of an expected increase of rigidity for soils. Considering the absence of water table in the ground due to high slope and permability, this correlation is observed at both daily and seasonal time scales, with a maximal amplitude of +/- 3%, and a very short response delay (several hours only). We precisely describe and quantify these correlations, towards in a second time being able to correct reversible temperature effects on dV/V and distinguish them from other possible processes precursor to failure. As temperature seems to have a stabilisation effect on slope due to an increase of the ground stiffness (dV/V) with temperature, our study paves the way to investigate more in details thermo-elastic processes on landslides as already done for rock columns (Guillemot et al., 2022).   Keywords: seismic interferometry, dV/V, temperature, landslide, monitoring