Developing models to prevent landslides



May 2010. A home in Saint-Jude, just outside Saint-Hyacinthe, is swept away in a landslide. A couple and their two daughters perish, buried under metres of mud. To prevent a similar tragedy, Ariane Locat, professor in the Department of Civil Engineering and Water Engineering at Université Laval, is working with the Ministère des Transports, de la Mobilité durable et de l'Électrification des transports du Québec and the Ministère de la Sécurité publique du Québec to demystify these types of major landslides, known as spreads, that occur in eastern Canada's sensitive clay soils and impact areas spanning over one hectare.

Spreads happen fast and without warning according to a process that remains poorly understood.

Spreads happen fast and without warning according to a process that remains poorly understood. The researcher is exploring the hypothesis that the ground breaks up as a result of a progressive fracture caused by erosion or human intervention that starts at the foot of a slope and stretches horizontally. In other words, under strain, the ground reaches its maximum resistance and buckles, creating a domino effect.

To test her theory, Professor Locat collects soil samples that she places in a ring shear tester adapted to study the behaviours of different types of clays. Calibrated according to data gathered from earlier spreads, the tester aims to reproduce the conditions observed in nature by applying a load to collapse the sample. Locat and her team will rely on their field and laboratory observations to develop models to explain the mechanisms that drive these devastating spreads and identify high-risk areas. Ariane Locat believes that her concept could also be adapted to sand liquefaction and avalanches.