PROBABILISTIC ANALYSIS OF GROUNDWATER LEVELS IN HILLSIDE SLOPES

Slope failures on hillside slopes are frequently caused by increases in pore-water pressures following periods of heavy rainfall. This paper presents a simplified lumped-parameter model for estimating piezometric levels due to infiltration of rainfall. The model requires a minimal number of parameters and is applicable to hillside slopes with a layer of soil over an impermeable base. Individual components are built in the model for evapotranspiration, infiltration, and drainage due to gravity flow. The model is used to estimate piezometric levels of small basins in the watersheds of Kennel Creek, Alaska, and Perkins Creek, Oregon. The estimated piezometric levels are compared with observed values, and the agreement is satisfactory. An uncertainty model based on the first-order, second-moment method is used to estimate the uncertainty about the moisture contents of the unsaturated zone and piezometric levels in terms of uncertainties about the input parameters. Bayesian updating is formulated and the observed piezometric levels are used to update the parameters.