Lawrence Berkeley National Laboratory

A common factor in landslide activation (or reactivation) is subsurface moisture and associated pore pressure variations linked to rainfall. Monitoring of these subsurface hydrogeological processes is necessary to improve our understanding of water-induced landslide activation. Geophysical approaches, electrical methods in particular, are increasingly being applied to landslide monitoring because they provide non-invasive spatial information in heterogeneous subsurface environments that can be difficult to characterise using surface observations or intrusive sampling alone. Electrical techniques are sensitive to changing subsurface moisture conditions, and have proven to be a useful tool for investigating the hydrogeology of natural and engineered slopes. The objectives of this investigation were to further develop electrical resistance monitoring for slope stability assessment, and to validate the approach at an intermittently-active UK landslide system to advance the understanding of complex landslide activation mechanisms. A long-term transfer resistance dataset was collected from a grid of electrodes to allow spatial monitoring of the landslide. These data were interpreted using a synthesis of rainfall, temperature, GPS and piezometric records. The resistance data were corrected for seasonal temperature variations and electrode movements were monitored, as these processes were shown to mask moisture related changes. Results reveal that resistance monitoring is sensitive to soil moisture accumulation, including changes in piezometric levels, and can be used to study the principal activation mechanism of slow-moving shallow earthflows. Spatial monitoring using resistance maps was shown to be particularly valuable as it revealed the evolution of subsurface moisture distribution, in the lead up to landslide activation. Key benefits of this approach are that it provides a simple, rapid and non-invasive means of spatially monitoring subsurface moisture dynamics linked to landslide activation at high-temporal resolution. Crucially, it provides a means of monitoring subsurface hydraulic changes in the build-up to slope failure, thereby contributing to early warning of landslide events. Copyright © 2018 John Wiley & Sons, Ltd.