WHAT CAN LANDSLIDE DEPOSITS THROUGHOUT THE PACIFIC NORTHWEST TELL US ABOUT CASCADIA SUBDUCTION ZONE EARTHQUAKES?

While a wealth of onshore and offshore geological data continue to improve our understanding of Cascadia Subduction Zone (CSZ) earthquake recurrence, comparatively little observational data are available to constrain the severity of shaking during megathrust earthquakes. Records of past CSZ earthquake ground motions are limited to scattered historical accounts that describe strong shaking across the margin, but few observations of liquefaction, landslides, or other geologic evidence are available to model past shaking intensities. These limited data have led to a wide range of estimated paleoshaking intensities in western Oregon, from very strong (>1 g) to relatively weak (<0.25 g). Here, we synthesize the results of two studies in coastal Oregon that combine dendrochronology, geomorphic proxies, and numerical slope stability modeling to 1) place bounds on the rates of landslide triggering during past CSZ earthquakes and 2) determine minimum shaking intensities necessary to trigger a subset of those landslides.

In the first study, we leverage two datasets of landslide numeric ages to determine the range of coseismic landslide triggering rates in Central Oregon: a set of high-precision dendrochronologic ages of landslide-dammed lakes, and another with nearly 10,000 estimates of landslide triggering ages from age-calibrated surface roughness. Our results constrain the range of Cascadia earthquake landslide triggering rates to 0 – 12% of all landslides preserved in the modern Oregon Coast Ranges. In our second study, we map landslides along marine terraces of the Oregon coast and used 3D slope stability modeling to identify landslide morphologies that could only be triggered by strong earthquake shaking. Using the 222 landslides we identify as coseismic, we invert landslide geometry to solve for the minimum shaking required for failure. This analysis yields values of peak ground acceleration comparable to recent model estimates for full-margin ruptures. This suggests that coastal terrace landslides may serve as an effective recorder of past strong ground motion. Taken together, these studies provide a novel along-strike measure of Cascadia earthquake paleoshaking intensities derived from the morphology of coastal landslides and help constrain a broad region of coseismic landslide hazard in Cascadia.