A new type of sound sensor system has been developed to predict the likelihood of a landslide. Thought to be the first system of its kind in the world, it works by measuring and analysing the acoustic behaviour of soil to establish when a landslide is imminent so preventative action can be taken.Noise created by movement under the surface builds to a crescendo as the slope becomes unstable and so gauging the increased rate of generated sound enables accurate prediction of a catastrophic soil collapse.

The technique has been developed by researchers at Loughborough University, in collaboration with the British Geological Survey, through two projects funded by the Engineering and Physical Sciences Research Council (EPSRC).The detection system consists of a network of sensors buried across the hillside or embankment that presents a risk of collapse. The sensors, acting as microphones in the subsoil, record the acoustic activity of the soil across the slope and each transmits a signal to a central computer for analysis.

Noise rates, created by inter-particle friction, are proportional to rates of soil movement and so increased acoustic emissions mean a slope is closer to failure. Once a certain noise rate is recorded, the system can send a warning, via a text message, to the authorities responsible for safety in the area. An early warning allows them to evacuate an area, close transport routes that cross the slope or carry out works to stabilise the soil.

Neil Dixon, professor of geotechnical engineering at Loughborough University and principal investigator on the project, explains how the system – thought to be a global first – works. “In just the same way as bending a stick creates cracking noises that build up until it snaps, so the movement of soil before a landslide creates increasing rates of noise,” said Professor Dixon.”This has been known since the 1960s, but what we have been able to do that is new is capture and process this information so as to quantify the link between noise and soil displacement rates as it happens, in real time – and hence provide an early warning,” he added.