Evidence of Dilatant Strengthening as a Mechanism Controlling Landslide Velocity
Abstract
The types of hazards posed by landslides are largely determined by landslide velocity, but mechanisms controlling velocity are not fully understood. To evaluate these mechanisms, we made hourly measurements of displacement, pore-water pressure, and meteorological conditions for four years at the continuously moving Slumgullion landslide in southwestern Colorado, USA. We performed episodic topographic surveys and in-situ and laboratory tests to estimate physical properties of landslide materials. Monitoring revealed that the average landslide velocity was 1.0 cm/d with maximum velocity reaching 1.7 cm/d during spring snowmelt and intense and prolonged rainfall events. Acceleration was contemporaneous with elevated pore pressures away from the shearing landslide margin. Along the margin, however, acceleration was accompanied by decreased pore pressures, suggesting dilation of shearing material at rates greater than rates of pore-fluid movement. Our hypothesis of dilation induced by shearing was supported by topographic survey results and density measurements made across part of the landslide margin; soil densities were lowest and the ground surface had relative upward movement near where shearing occurred. Ring-shear tests of several landslide samples verified that shearing induces dilation in this material (dilatancy averaged 3.2 x 10-4). From our observations, we estimate that the rate of pore opening (product of velocity and dilatancy) averaged 3.2 x 10-4 cm/d and had a maximum of 5.4 x 10-4 cm/d. In-situ and laboratory measurements of hydraulic conductivity provided values of 3.6 x 10-3 (near the active shear) to 4.8 x 10-5 cm/d (across the shear zone). Therefore, because the hydraulic conductivity measured across the shear zone is an order of magnitude lower than the rates of pore opening, we conclude that dilation generally occurs at rates greater than pore fluid movement, resulting in dilatant strengthening of landslide material and reduction of landslide velocity. However, we found that shear-induced dilation only occurred during the initial 20-1,100 cm of shear displacement, indicating that landslide material cannot be continuously sheared and still dilate. We have observed relocation of the actively shearing landslide margin within series of flank ridges and propose that shear displacement is temporary at given locations along the base and margins of the landslide. Relocation of active shearing probably results from changing stress conditions within the landslide and triggers episodic shear-induced dilation and consolidation of landslide material. Consolidation of formerly sheared material and shearing of previously unsheared material allows dilatant strengthening to occur along many parts of the shearing boundary of the landslide at any given time. Dilatant strengthening can, therefore, continuously regulate landslide velocity regardless of shear displacement magnitudes. Recognition of this mechanism controlling landslide velocity should aid assessments of landslide hazards.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2008
- Bibcode:
- 2008AGUFM.H43I..03S
- Keywords:
-
- 1810 Debris flow and landslides;
- 1822 Geomechanics;
- 1826 Geomorphology: hillslope (1625);
- 1829 Groundwater hydrology;
- 1895 Instruments and techniques: monitoring