Laboratory experiments on the influence of soil pipes on slope failure

Abstract

Existence of soil pipes is reported in many collapsed slopes indicating their influences on landslide initiation. Flume model tests with different soil pipe configurations, (a) no pipe, (b) closed pipe and (c) open pipe, were conducted to understand the influence of soil pipes on slope stability during rainfall. Porewater pressures, discharges through the soil pipe and seepage from the slope end were recorded. Open pipe works as a hillslope drainage and reduces the porewater pressure of an entire slope, but if blocked, porewater pressure close to the lower end of pipe rises up rapidly, leading to immediate soil mass movement. Average hillslope saturation at the time of failure in a hillslope with “initially open but later blocked” pipe is lower than that in other cases. Discharge measurement shows that soil pipe even if ended within the hillslope, increases hillslope discharge. In all the three cases of experiments, shallow retrogressive sliding occurred. However, backward progress is faster in the hillslope without soil pipe because of more uniform distribution of moisture throughout the slope than the other two cases. The simple numerical analysis developed is able to predict the timing of the first crack for piped and no-piped hillslope.