2009 Portland GSA Annual Meeting (18-21 October 2009)

Paper No. 3
Presentation Time: 2:05 PM

RAPID RESPONSE OF DEEP BEDROCK LANDSLIDES TO RAINFALL, COASTAL OREGON


SCHULZ, William H., U.S. Geological Survey, MS 966, Box 25046, Denver, CO 80225, GALLOWAY, Sarah L., U.S. Geol Survey, MS 966, Box 25046, Denver, CO 80225 and HIGGINS, Jerry D., Dept Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401, wschulz@usgs.gov

Numerous deep, translational landslides occur in Tertiary sedimentary rock along the central coast of Oregon and pose hazards to human safety and public and private property. Different from most deep landslides, these landslides reactivate during or soon after heavy rainfall. The mechanisms responsible for episodic reactivation of the landslides must be understood to reduce the hazards posed by their movement. To increase our understanding, we are studying three of the landslides in detail. Our studies include engineering geologic mapping, subsurface exploration, rock and soil sampling, and nearly continuous monitoring of landslide displacement and meteorological and hydrological conditions. The landslides extend from 15-30 m high coastal bluffs landward into the adjacent gently sloping, former marine terrace. The slides have lengths of 50-200 m and widths of 200-550 m. The maximum depth of the landslides is 30 m and basal rupture surfaces slope gently with maximum inclination of about 20°. The water table within the slides also slopes gently seaward and has an average depth of about 15 m. Groundwater flow is generally parallel to the water table and seaward. Two of the landslides studied moved several times during the winter of 2008/09 and did so nearly simultaneously, while the third also moved but with unknown timing because of an equipment malfunction. Observed movement occurred when pore-water pressures exceeded consistent thresholds and within hours of commencement of heavy rainfall. We observed that pore-water pressures increased simultaneously at all saturated depths below a given location on the landslides. Increases occurred first near the landslide heads and propagated laterally toward the landslide toes. Only near the landslide heads was infiltration of rainfall to the water table observed in advance of pore-water pressure increases below the water table. Our findings suggest that mitigation efforts could be directed toward reducing infiltration near and upslope from the landslide heads, and that reactivation of similar landslides in the region may be forecast using pore-pressure thresholds and monitoring results from only one landslide.