A large scale sprinkling experiment on the effects of the entrapped air on the land-slide initiation

To examine the effects of entrapped air for the land slide, a large-scale sprinkling experiment was conducted. Previous studies on runoff generation in hillslopes entrapped air can cause for the pressure head increasing and enhancing runoff generation (Onodera 1991). Since entrapped air in sewer system sometimes blew off the manhole (Haman and McCorquodale 1982), entrapped air in fractured bedrock can cause of the large force to move the soil mass, resulting in landslides. The purpose is to study the influence of the entrapped air on the land-slide, by using large-scale rainfall simulator in the National Research Institute for Earth Science and Disaster Prevention in Tsukuba, Japan. The experimental slope with 6.25 m long, 1.5 m wide and 30 degrees has an air pocket with 1 m long, 1 m wide and 10 cm height middle part of slope. We monitored soil suction by tensiometers, runoff and entrapped air pressure. Air was injected from the hole at the bottom of the slope through the air compressor. In order to analyze the soil suction response by injected air, 0.8 MPa air was injected for 3 minutes in 100 mm h-1 rainfall. The distinct change for discharge was observed during the air injection, and rotational slip with about 2 m long, and about 30 cm max depth was observed. The deep tensiometers indicate marked increase at the same time, but the shallow tensiometers did not respond to the air injection, and entrapped air pressure perturbation ware less than 2 kPa. Marked increase of hydraulic gradient was observed at the time of the rotational slips. This suggests that this increase of hydraulic gradient should be the cause of increasing runoff by the air injection. Therefore perturbation of entrapped air pressure would become trigger of landslide, by increasing hydraulic gradient of the groundwater.