Time-Prediction Method of the Onset of a Rainfall-Induced Landslide Based on the Monitoring of Shear Strain and Pore Pressure

Abstract

It is important to simulate the shear deformation of a slope under rainfall to predict the onset of rainfall-induced landslides. Monitoring of deformation and soil–water characteristics in a sandy slope model under artificial rainfall was conducted to establish a prediction method for shear deformation of the slope due to rainfall infiltration and the onset of a rainfall-induced landslide. A hyperbolic relationship between the shear strain and the pore pressure at the same depth was identified from the analysis of the monitored data. A time-prediction method of the shear strain in the slope was established based on the relation as follows. Regression analyses of the shear strain—the pore pressure relationship at any given time before the failure of the slope—and the time—the pore pressure relationship at the same time—were conducted first. Combining both equations produced an equation for the relationship between the time and the shear strain. The equation simulated the time variation of shear strain in the slope relatively well. Then, an equation for the relationship between the time and the inverse of the shear strain velocity was derived by differentiating the equation for the relationship between the time and the shear strain. This time-prediction method predicts the failure time relatively well, especially at deeper soil layers in the slope when the shear deformation proceeds with the smooth increase of positive pore pressure. The shear strain cannot follow rapid jumping up of the pore pressure. The results showed that the procedure described above could be applicable for the prediction of the time variation of shear deformation and the failure time of the slope.