ASCE has honored Jordan Aaron, Ph.D.; Oldrich Hungr, Ph.D.; Timothy Stark, Ph.D., P.E., D.GE, F.ASCE, and Ahmed Baghdady, Ph.D., A.M.ASCE, with the 2019 Norman Medal for their paper “Oso, Washington, Landslide of March 22, 2014: Dynamic Analysis,” Journal of Geotechnical and Geoenvironmental Engineering, September 2017.
This paper provides a means for predicting the runout distance of future landslides with similar topography and failure mechanism by creatively modeling the behavior of the water-filled colluvium, which traveled 1.5 km. The resulting runout analysis can be used to identify the hazard and assess the risk, i.e., probability of runout multiplied by the potential loss, of future slope failures to protect life and property. If the analysis presented in this paper (described below) had been performed before the Steelhead Community was permitted, it would have been clear that the runout for a high-elevation landslide would pass through the community and well past SR-530.
The proposed method is also applicable to the design and risk assessment of natural and man-made earthen slopes that are near residential and commercial developments. The proposed runout analysis can be used to predict the runout distance of future landslides and slope failures to assess the risk to new and existing infrastructure below the slope. The important finding and contributions of this paper are—
- The Mar. 22, 2014, landslide occurred in two phases, with Phase I impacting, pushing, and overriding the waterfilled, disturbed, and softened colluvium along the slope toe from prior low-elevation landslides, causing a dramatic undrained strength loss (liquefaction) that enabled the colluvium to flow approximately 1.5 km across the valley and kill 43 people.
- The Phases II landslide remained on the slope and did not undergo a large strength loss. Thus, only the material derived from previous failures liquefied.
- The novel concept in this paper is that an overconsolidated glacial clay could liquefy and the resulting undrained strength can be approximated using existing empirical liquefied strength correlations that were developed using earthquake-induced liquefaction cases of sandy materials.
- This is important because a number of runout landslide models exist (DAN3D, Flo2D, ANURA3D, etc.) but guidance on estimating appropriate input parameters was not available before this paper, so predicted runout models were not accurate.
- The liquefied strength of the water-filled, disturbed, and softened colluvium along the slope toe was modeled using a liquefied strength ratio 0.07 estimated from an estimated 33 earthquake-induced liquefaction flow failures in loose, saturated granular materials because they exhibited similar behavior. The resulting runout distance and geometry are in excellent agreement with the observed runout, which is well past the Steelhead Community and SR-530.
- The resulting runout analysis improves landslide detection, prediction, hazard mapping, land-use planning, and property rights policy decisions by better predicting situations where large and mobile landslides can occur. This will result in a better ability to develop reasonable setback distances from top and bottom of slopes and provide a more objective assessment for land use and property rights issues. The proposed runout model can also be used to assess various remedial measures in a quantitative manner to determine appropriate ways to safeguard the public and property.
The Norman Medal is bestowed upon the author or authors of a paper that is judged worthy of special commendation for its merit as a contribution to engineering science.
