9:15 AM - 9:30 AM
[HGM03-04] Formation of sliding surface in rainfall-induced shallow landslides:
focusing on the subsurface hydraulic structure and groundwater percolation processes in soil and bedrock
Keywords:percolation, permeability, soil layer, hillslope, shear failure
To reveal the conditions under which rainfall-induced shallow landslides occur, we have to reveal what kind of hydrological processes triggers shear failure. At depths where permeability changes drastically, such as at the boundary between soil layer and weathered bedrock, not all the infiltrated water can percolate vertically, and shear failure occurs due to an increase in pore water pressure. However, at depths where permeability does not decrease significantly, such as inside the soil layer, the water movement that causes shear failure has not been revealed clearly. Here, we focused on granitoids hillslopes in the south of Miyagi Prefecture where many shallow landslides occurred due to a heavy rainfall event in 2019, observed the change of pressure head in the shallow depth from ground surface using tensiometers and revealed hydrological processes before shear failure by spatial projection of hydraulic head.
Firstly, we selected one hillslope each from granite and granodiorite bedrocks. The selected hillslopes are located in zero-order basins and besides nearby the scar head but retain soil layer. The former is an example where the boundary between the soil layer and the weathered bedrock corresponds with the slip surface, and the latter is an example where the slip surface corresponds with the depth where the permeability in the soil layer changes slightly towards the vertical direction. Secondly, we placed monitoring points at intervals of about 1.5 m along the maximum slope direction and installed 3-6 tensiometers/monitoring point (25 tensiometers in total). One rain gauge was installed in the depositional area of the granodiorite site. We monitored from May to November 2021. Finally, we focused on the rainfall event at the end of July 2021, calculated the hydraulic head of the monitoring points, interpolated linearly between the monitoring points, and spatially projected the change in hydraulic head every 10 minutes. The target rainfall event is characterized by a heavy rain after a slightly weak rain, which is similar to the characteristic of the heavy rainfall event in 2019.
The calculated results indicate that not all the infiltrated water can percolate vertically at depths where the permeability changes slightly towards the vertical direction. In the granite hillslope, the results show that infiltrated water percolates quickly and vertically, and that not all the infiltrated water can percolate vertically at the boundary between the soil layer and the weathered bedrock, resulting in saturated throughflow. On the other hand, in the granodiorite hillslope, the results show that not all the infiltrated water can percolate vertically at the boundary between the upper and lower soil layer, and subsurface water stagnates at the boundary. Based on the above, we will discuss whether the depths at which hydraulic conductivity decreases by about 10-1 (cm/s) towards the vertical direction are potential sliding surfaces, regardless of the boundary between upper/lower soil layer or soil layer/weathered bedrock.
Firstly, we selected one hillslope each from granite and granodiorite bedrocks. The selected hillslopes are located in zero-order basins and besides nearby the scar head but retain soil layer. The former is an example where the boundary between the soil layer and the weathered bedrock corresponds with the slip surface, and the latter is an example where the slip surface corresponds with the depth where the permeability in the soil layer changes slightly towards the vertical direction. Secondly, we placed monitoring points at intervals of about 1.5 m along the maximum slope direction and installed 3-6 tensiometers/monitoring point (25 tensiometers in total). One rain gauge was installed in the depositional area of the granodiorite site. We monitored from May to November 2021. Finally, we focused on the rainfall event at the end of July 2021, calculated the hydraulic head of the monitoring points, interpolated linearly between the monitoring points, and spatially projected the change in hydraulic head every 10 minutes. The target rainfall event is characterized by a heavy rain after a slightly weak rain, which is similar to the characteristic of the heavy rainfall event in 2019.
The calculated results indicate that not all the infiltrated water can percolate vertically at depths where the permeability changes slightly towards the vertical direction. In the granite hillslope, the results show that infiltrated water percolates quickly and vertically, and that not all the infiltrated water can percolate vertically at the boundary between the soil layer and the weathered bedrock, resulting in saturated throughflow. On the other hand, in the granodiorite hillslope, the results show that not all the infiltrated water can percolate vertically at the boundary between the upper and lower soil layer, and subsurface water stagnates at the boundary. Based on the above, we will discuss whether the depths at which hydraulic conductivity decreases by about 10-1 (cm/s) towards the vertical direction are potential sliding surfaces, regardless of the boundary between upper/lower soil layer or soil layer/weathered bedrock.