Cordilleran Section - 99th Annual (April 1–3, 2003)

Paper No. 2
Presentation Time: 8:30 AM-5:30 PM

DYNAMICS OF THE SWIFT CREEK LANDSLIDE, NORTHWESTERN WASHINGTON


MCKENZIE-JOHNSON, Alexander S. and CLARK, Douglas H., Geology, Western Washington Univ, 516 High Street, Bellingham, WA 98225-9080, Alexandermj@hotmail.com

Deep-seated landslides are common yet poorly understood agents of landscape evolution in Washington State. The Swift Creek landslide in northwestern Washington is an active, accessible, and particularly well-developed example of such landslides. Our study investigates modern and historical movement of the landslide by combining a detailed GPS topographic survey, repeated surveys of 38 distributed benchmarks, mapping of surficial features, tree-ring analysis of tilted trees (dendrogeomorphology) and interpretation of historical aerial photographs.

Modern activity on the landslide appears to have begun in the 1930s. Dendrogeomorphology and historical aerial photographs suggest that the south lateral margin had an apparent average horizontal velocity of ~5 m/y between 1955 and 2002. The landslide does not appear to have changed dramatically in size since the 1940s. However, the lower 1/3rd of the landslide (zone of accumulation) appears to have thickened substantially since that time, and is apparently being fed by material derived from thinning of the upper 2/3rd (zone of depletion) of the landslide, rather than by headward expansion of the landslide.

To assess the short-term dynamics of the landslide, we established thirty-eight monitoring benchmarks across the landslide in July of 2002 and surveyed their locations with Differential-GPS (Trimble Pro-XR; max. horizontal resolution of 0.2 m). Our initial resurvey of the benchmarks (mid-September, 2002) indicates rates of displacements of between 0.42-4.45 cm/day (~1.5-16 m/yr), with the majority of movement roughly parallel to the regional surface slope. The highest rates of movement occur along the north (right-lateral) side of the toe, and adjacent to the north lateral scarp, whereas the lowest velocities occur at the base of the toe. Preliminary indications are that the landslide is composed of a series of rotational blocks at the head of the landslide transitioning to lateral movement at the toe of the landslide. Multiple resurveys of the landslide during the coming winter and spring will help refine these preliminary results.