GSA Annual Meeting in Denver, Colorado, USA - 2016

Paper No. 277-10
Presentation Time: 10:45 AM

APPLICATION AND TESTING OF A COUPLED HYDROMECHANICAL MODEL OF HILLSLOPE HYDROLOGIC RESPONSE AND SLOPE STABILITY, MUKILTEO, WA


MIRUS, Benjamin B., U.S. Geological Survey, Geologic Hazards Sciences Center, Golden, CO 80401, LU, Ning, Civil Engineering, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, WAYLLACE, Alexandra, Civil and Environmental Engineering, Colorado School of Mines, 1500 Illinois St, Golden, CO 80401, GODT, Jonathan W., U.S. Geological Survey, Denver Federal Center, P.O. Box 25046, MS 966, Denver, CO 80225, SMITH, Joel B., U.S. Geological Survey, Geologic Hazards Sciences Center, Denver Federal Center, P.O. Box 25046, MS 966, Denver, CO 80225 and BAUM, Rex L., U.S. Geological Survey, Geologic Hazards Science Center, MS 966, Box 25046, Denver, CO 80225, bbmirus@usgs.gov

Prolonged or intense rainfall in steep terrain can lead to shallow landslide initiation. Quantitative assessments of shallow landslide hazards often rely on simplified numerical models of steady infiltration coupled indirectly with slope stability assessments that treat the failure prone portions of the hillslope as a uniform “infinite” slope. Improved quantitative characterization of when and where shallow landslides may initiate likely requires a more robust physically-based representation of the interaction between the hydrologic and geomechanical process leading to slope instability. Here we explore a coupled hydrogeomechanical modeling approach for assessment of temporally variable hydrologic response and slope stability within a hillslope. The approach uses HYDRUS2D, a two-dimensional Richards’ equation based model for subsurface flow through variably saturated porous media, coupled with the Slope Cube module, which uses simulated hydrologic response to directly calculate the local factor of safety field across the hillslope. We apply the coupled model to a pair of coast bluff hillslopes located in the City of Mukilteo, Washington. One hillslope is densely vegetated and currently stable whereas the other is an active landslide that fails periodically through a series of translational slides. We parameterized the simulations for these two hillslope transects using detailed site characterization and laboratory analyses of specimens collected from each site. We evaluated the simulations using continuous time-series of measured volumetric water content, pore pressures, and shallow water-table fluctuations as well as the timing (or absence) of recorded slope movements. Simulation results for both hillslopes highlight the important role of rainfall dynamics, colluvium thickness, topographic variations along the hillslope transects, and apparent cohesion from tree roots in controlling hydrologic response and slope stability.