PRELIMINARY ASSESSMENT OF POST-FAILURE BEHAVIOR OF GLACIOLACUSTRINE SILTY CLAY FROM THE 2014, SR530 (OSO), WASHINGTON LANDSLIDE

In 2014, a highly mobile landslide near Oso, Washington tragically killed 43 people. Glaciolacustrine silty clay (“clay” herein) with ~40% clay-fraction variability formed the lower half of the landslide. Landslides in clay are common but movement ranges from slow and episodic to rapid and catastrophic. Although long studied, some important questions regarding shear failure of clay remain unanswered. To improve understanding, we used a large ring-shear apparatus to test 2 samples of clay from the landslide under Skempton’s fully softened and residual shear strength conditions. Ring-shear devices are most appropriate for studying large displacements and variable shear speeds. We tested the siltier part of the glaciolacustrine unit.

Tests indicated that shear strength dropped by about half during the initial ~30 cm of shear displacement when initially fully softened. Strength loss generally was accompanied by increased pore-water pressure but in the absence of significant volume change; a test at lower normal stress displayed decreased (negative) pore pressure. Atypically, ~28% strength loss with displacement occurred when sheared at residual strength and strength also decreased with increased shear speed. Residual shear zones dilated during displacement and developed negative pore pressures at low normal stress but positive pressures at high normal stress. The more clayey of the two specimens exhibited lower fully softened and residual strengths. Post-test inspection revealed better-developed slickensides for the more clayey specimen and that shear zones were finer grained, had lower porosity, and higher water content than the remainder of each sample. Physicochemical mechanisms are known to affect clay strength. Our observations support this, with shear strength variability correlating poorly with changes in volume and pore pressures but correlating well with clay-sized fraction.

Our preliminary results compared to those from studies of less mobile landslides suggest that observed strength loss was insufficient to fully explain the landslide’s mobility. Detailed geomorphological reconstruction of the landslide event and tests on other materials, including more clayey parts of the glaciolacustrine unit, are needed to better understand mechanisms contributing to the landslide’s mobility.