Executive summary

The stretch of the A83 road that passes through the Rest and Be Thankful is subject to a chronic and persistent landslide risk. This location is crucial, both as a tourist stop and as a significant pinch point that provides access to the western region of Scotland.

It is significantly impacted by a high frequency of landslides, which typically occur in association with heavy rainfall, but as yet direct associations between slope failures and rainfall thresholds have not been consistently reliable. Part of the reason may be associated with the propagation of rain and groundwater through the slope material.

Stakeholders have often noted precursory deformation in the area, which sometimes develops into debris flows that threaten road operations. In addressing the challenge of detecting landslides, we have employed both visual and passive seismic continuous monitoring approaches.

The visual method involves the use of Particle Image Velocimetry (PIV) applied to near real-time streamed time-lapse photographs, while the new passive seismic monitoring is carried out using ground movement sensors. PIV involves analysing the movement of particles (pixels) from time-lapse photographs to determine the magnitude and direction of slope movements, providing a unique chronology of progressive slope movement.

We have worked to make this operational by adding the functionality to automatically alert designated operators when change is detected from the most recent image. However, dedicated operators are required to set locally specific thresholds and areas of interest, to ensure the processing is running, and to validate outputs.

The passive seismic approach has been demonstrated to potentially detect and even locate events, but the continuous data streaming that is required limits its operational feasibility with the current infrastructure and signal strength at the site. The best option would be to further develop the system for in situ processing and only stream alerts if a set threshold was detected, which would require an in-depth study of the specific signals relative to ambient noise.

The passive seismic system also holds exciting and demonstrable potential to detect changes in ground moisture conditions that may be a more direct indicator of critical slope conditions than rainfall, with far lower power and communication links required to achieve the desired data.

The use of passive seismic monitoring in combination with co-located electroseismic sensors is introduced as a powerful new combination for the monitoring of the near-surface water saturation.