GSA Annual Meeting in Seattle, Washington, USA - 2017

Paper No. 216-12
Presentation Time: 11:30 AM

SOIL SWELLING INHIBITS LANDSLIDE INITIATION AND ACCELERATION


SCHULZ, William H., U.S. Geological Survey, MS 966, Box 25046, Denver, CO 80225, SMITH, Joel B., U.S. Geological Survey, Geologic Hazards Science Center, Denver Federal Center, P.O. Box 25046, MS 966, Denver, CO 80225, WANG, Gonghui, Research Center on Landslides, Kyoto University, Gokasho, Uji, Kyoto, 611-0011, Japan, JIANG, Yao, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu, 610041, China and ROERING, Joshua J., Department of Geological Sciences, University of Oregon, 1272 E. 13th Ave, Eugene, OR 97403, wschulz@usgs.gov

Landslides present significant hazards and importantly modify landscapes, both of which are affected by the timing of landslide initiation and consequent landslide mobility. We studied in detail a slowly moving landslide (earthflow) typical of thousands in northern California to improve understanding of landsliding mechanisms. Continuous monitoring during November 2014–July 2017 revealed three periods of motion, two of which occurred following the rainy season, and one that occurred during a relatively normal rainy season. Groundwater pressure head measured at six locations within the landslide varied annually by ~3 m. We observed 3 cm of seasonal swelling of near-surface landslide material and concurrent swell pressure generation that exceeded 25 kPa, whereas laboratory tests indicated inherent swell potential for landslide material as high as 695 kPa. Ring-shear strength testing indicated essentially rate-invariant residual strength of landslide boundaries. Landslide movement timing and speed correlated poorly with observed groundwater pressures, with movement initiating at inconsistent pressures and peak speeds occurring after peak heads had passed. Limit-equilibrium slope-stability modeling using measured strength and hydrological conditions corroborated these findings, also suggesting that landslide initiation should have occurred ~5–6 months prior to when it was observed. However, when a measure of transient swelling was added to shear resistance, modeling results well replicated observed landslide timing and speed. Infiltration of precipitation causes soil swelling that progresses to greater depth with time, which increases resistance to landslide movement along shear zones bounding the landslide’s sides, much like tightening of a caliper brake on a wheel. Eventually, elevated groundwater pressure along bounding shear zones may overcome the swell-induced braking and instigate motion. By contrast, elevated pore pressure during the rainy season may be insufficient for instigating motion until shrinkage of near-surface material occurs, much like releasing the brakes on the wheel. Our findings suggest that swell pressures fundamentally control landsliding in clayey soils and thus warrant consideration in evaluations of landslide timing and mobility.