Observed and modeled rainfall conditions for shallow landsliding in the Seattle, Washington, area

Shallow landsliding on steep hillslopes in the Seattle area has caused the loss of human life and millions of dollars in damage to private property and public infrastructure. Widespread shallow landsliding is typically associated with moderately intense rainfall of long duration on already wet hillslope materials. The effective management of landslide hazards requires knowledge of the spatial and temporal conditions that lead to shallow slope failure. The likelihood that any location will fail as a shallow landslide is a function of the topographic slope and the hillslope material strength modified by the transient effects of infiltration and groundwater. I examined the rainfall and soil moisture conditions that triggered widespread shallow landslides in Seattle by comparing hourly rainfall information with a complementary database of landslide locations and dates. Results show that rainfall on already wet hills lope materials in excess of a threshold given by I = 82.73D-1.13, where I is rainfall intensity in mm/hr and D is rainstorm duration in hours, has caused widespread shallow landsliding six times during the period 1983 to 1997. I then used a numerical solution to the governing equations of transient groundwater flow to simulate the pore-pressure response to realistic rainfall. The hydraulic parameters of hillslope colluvium were determined using capillary-rise tests. Model output was compared to field monitoring data to assess the accuracy and appropriateness of the input data. Results were cast in terms of a destabilizing pore-pressure response for several hypothetical hillslopes and rainstorms. Finally, I applied a distributed, transient infiltration - slope stability model to an 18-km2 area of southwestern Seattle. The model combines an infinite slope-stability calculation and an analytic, one-dimensional solution for pore-pressure diffusion in a soil layer of finite depth in response to time-varying rainfall. Material strength and hydraulic properties used in the model were determined from field and laboratory measurements and a saturated initial condition was assumed. Because the solution to the groundwater-flow equations explicitly account for the response of pore pressure at depth to a flux at the ground surface with respect to time, results can be portrayed quantitatively to assess the potential landslide hazard based on rainfall conditions.