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

Paper No. 35
Presentation Time: 9:00 AM-6:00 PM

ANCIENT LANDSLIDES, LANDSLIDE DAMS, AND OUTBURST FLOOD POTENTIAL IN AN EXTENSIONAL SETTING


CROALL, Kelsey, Environmental Studies, Lewis & Clark College, 0615 SW Palatine Hill Road, Portland, OR 97219, JONES, Emily, Sociology/Anthropology, Lewis & Clark College, 0615 SW Palatine Hill Road, Portland, OR 97219, SAFRAN, Elizabeth, Lewis & Clark College, 0615 SW Palatine Hill Road, Portland, OR 97219, O'CONNOR, Jim E., U.S. Geological Survey, 2130 SW 5th, Portland, OR 97216, HOUSE, P. Kyle, Nevada Bureau of Mines and Geology, University of Nevada, MS 178, Reno, NV 89557 and ELY, Lisa L., Dept. of Geological Sciences, Central Washington University, 400 E. University Way, Ellensburg, WA 98926, kcroall@lclark.edu

Large landslides can form valley blockages lasting from minutes to thousands of years. Relatively stable dams can inhibit incision upstream through aggradation, while unstable dams may promote incision locally via catastrophic floods. Previous empirical studies of landslide dams, primarily in rugged terrain, have used dam, reservoir, and basin morphometry to characterize propensity for abrupt dam failure. Here, we analyze a population of 413 large (0.1 to >40 km2) landslides in eastern Oregon -- a back-arc setting affected by Cenozoic volcanism, sedimentation, and extension. Scanty geochronologic data suggest landslide ages of order 104 to 105ka, and channels have re-established through or around all but one of the landslide deposits. Morphologic evidence from topographic data and imagery was used to identify possible former blockages. Though a comprehensive field survey has not been conducted, we have field evidence for ~10-15 blockages in the Owyhee, Deschutes, and John Day River basins. Approximately 30% of all mapped landslides impinge on channels with drainage areas between 10 and 100 km2, while 40% may have blocked channels of drainage area >100 km2. Because of their potential to affect upstream portions of the channel networks, potential blockages in relatively large channels were evaluated in further detail. The maximum height of a possible former blockage was assessed by finding the minimum alternative spillover elevation. Projecting this elevation upstream permitted estimation of the extent and volume of a temporary reservoir associated with a blockage. Additional estimations of blockage length and width allowed rudimentary computations of landslide dam volume. The mean blockage height of eastern Oregon landslide dams is at least 40% lower than the worldwide average, and estimated dam volumes are several-fold smaller, while estimated reservoir volumes are almost an order of magnitude larger, primarily due to lower regional gradients. The smaller ratio of landslide mass to water available for dam erosion makes the potential blockages of eastern Oregon appear highly unstable relative to other landslide dams populations, according to published indices. The impact of landslide dams on channels may therefore vary significantly with tectonic setting.