GEOLAB Transnational Access project RELERT - Risk and reliability study in quick clay

Quick clay is characterised by complete collapse and liquid-like mobility when overloaded. Quick clay is found primarily in Norway and Sweden, but also exists in Finland, Russia, Canada and Alaska. Quick clay landslides, with their retrogression characteristics and extreme mobility, pose significant risk to human lives, infrastructure, property and surrounding ecosystems. Hence, the proper characterization of quick clay sites is essential for ensuring the safety and resilience of infrastructure in Norway and elsewhere in Europe.

The current practice for mapping quick clay in Norway relies heavily on borehole data with either rotary sounding or total sounding and core samples tested in the laboratory. The only method for identifying quick clay with certainty is physical testing in the laboratory, but it is time-consuming, expensive and gives limited information, i.e., only at the depths and locations where the sample is taken. In Norway, rotary sounding and total soundings are frequently used in mapping of quick clay. In addition, geophysical methods such as Electrical Resistivity Tomography (ERT) are also occasionally used to supplement the results from soundings, particularly in early stage of ground investigation. Rotary sounding is regarded as the most cost-effective method for detecting potentially quick clay pockets.

The accuracy of quick clay mapping by a certain ground investigation method increases with increasing number of boreholes and decreasing distances to the boreholes in the area of interest. The accuracy is also increased with additional data sources such as ERT and/or other geophysical methods. Each rotary sounding, for example, provides information on the potential occurrence of quick clay with high accuracy at its location, but the accuracy decreases with increasing distance from the actual location of the rotary sounding. Current practice of mapping quick clay rarely deals with uncertainties associated with ground investigations quantitatively.  The uncertainties arise due to various sources including the interpretation of the rotary sounding, the number of soundings and the distance from the soundings, the equipment itself and the testing procedure. Dealing with uncertainty currently relies heavily on engineering judgement, which is subjective.

Transnational Access project RELERT is funded through the 1st call for proposal for the GEOLAB project. This project aims at developing models and validated on results from field testing to deal with uncertainty in ground characterisation in quick clay area more quantitatively.

The objectives of the RELERT project are:

(i) to obtain and process ground investigation data profiles at the Tiller-Flotten quick clay site (i.e. rotary soundings and ERT (electrical resistivity tomography) profiles. Notes: some of the data reported is obtained in earlier field investigation campaign but are processed and reported in this projects.

(ii) to build a mathematical model that characterizes the spatial variability of quick clays in Norway and

(iii) to develop a methodology for reliability assessment of quick clay sites under retrogressive failure.

We employ rotary soundings, soil samples and will use ERT-results in further study to learn the parameters of a hierarchical random field model, recently proposed by the user group for modelling the inherent spatial variability of material properties. The parameters of the model is learned by application of Bayesian analysis on the ground investigation data.

In further studies, we plan to use the model in slope reliability assessments under retrogressive slope failure through application of Monte Carlo methods. The results of the study are and will be documented in joint publications of the host and user groups.