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

Paper No. 12
Presentation Time: 9:00 AM-6:00 PM

THE 2009 RACEHORSE CREEK LANDSLIDE: FORENSIC DYNAMICS OF A LARGE, COMPLEX CATASTROPHIC MASS MOVEMENT


CRIDER, Juliet G., Department of Geology, Western Washington University, 516 High Street, MS 9080, Bellingham, WA 98225, TUCKER, David S., Mount Baker Volcano Research Center; Geology Department, Western Washington University, 516 High Street, Bellingham, WA 98225, CLARK, Douglas H., Geology, Western Washington Univ, 516 High Street, Bellingham, WA 98225 and LINNEMAN, Scott R., Geology Department, Western Washington Univ, 516 High St, Bellingham, WA 98225, juliet.crider@wwu.edu

The January 2009 Racehorse Creek landslide in Whatcom County, NW Washington is a composite, storm-triggered rock slide–debris flow, which released ~5x105 m3 of rock, much of which was delivered to Racehorse Creek. An “ear-witness” report indicates the failure occurred the afternoon of 7 January.

The landslide exposed a ~2 x104 m2 carbonaceous bedding plane (the main rupture surface, dipping 28° NW), within the Eocene Chuckanut Formation. We estimate the friction angle of this surface to be <27°. Simple limit equilibrium calculations suggest that this surface was unstable before the triggering rainstorms.

The kinematics of the landslide is complex, as evidenced by the spatial variation in striation trends on the failure surface and by the unusual distribution of deposits. The principal slip direction is NE, nearly perpendicular to the dip of bedding, and towards Racehorse Creek. The debris deposit in this direction comprises three lobes of large (≥ 4 m3) blocky debris. Absence of run-up on the opposite side of the canyon limits the velocity of the debris flow. A large secondary debris lobe flowed NW, bifurcating twice down slope. A third deposit at the base of the exposed slip surface is a ~5000 m2 translated block supporting disrupted forest floor.

The crown of the slide is at a local ridgeline; several en echelon, ~0.75 m-wide cracks in the crown show no significant change from May-August. Lateral release was enabled by bedding- and strike-perpendicular joints, creating a 27-m-high vertical wall at the S boundary. At the N edge of the exposed dip slope, a parallel weathered joint surface is exposed above the canyon.

Together, these observations suggest several components of failure for this landslide: Rainfall induced failure on existing joints at the edge of Racehorse Creek canyon. Retrogressive failure removed bedrock buttressing the main failure plane, releasing the main failure, which broke up and flowed/tumbled northward and down 240 m into the canyon. As the main source zone was evacuated, the higher zone slid down-dip on the low-friction bedding surface, transporting in situ stumps and forest floor into the original source area. From the highest elevation, debris gained enough momentum during translation to overtop a local ridge and form the NW deposits.