This thesis describes the laboratory modelling of gelifluction processes using the geotechnical centrifuge technique. Thirteen 1/10 scale planar slope models were frozen from the surface downwards on the laboratory floor and thawed, also from the surface downwards, under gravitational acceleration of 10 gravities (approximately 98.1 ms'2). A natural sandy silt soil from Quaternary periglacial slope deposits collected in SW England formed the base test material and slope models at gradients 4, 8, 12 and 16 were constructed using this soil. 10% and 20% by weight increments of glaciolacustrine silt and Kaolinite clay were added to the natural soil and a series of slope models were constructed at gradients of 4, 8, and 12 using these soils. Each slope model was subjected to four cycles of freezing and thawing except for the four slope models that underwent rapid slope failure. During thaw, soil temperatures and pore water pressures were recorded continuously, together with soil thaw settlement and surface displacement. Following each experiment, models were sectioned to observe displacement columns that showed the profiles of soil movement and allowed volumetric displacements to be calculated. It was shown that thaw settlement and slope gradient strongly affected the rate of surface movement and the subsurface profile of movement. Increasing slope gradient generated greater amounts of subsurface and surface movement as a function of increased gravitational shear stress. Thawing ice lenses inclined parallel to the slope gradient provided localised zones of microshearing in response to localised low frictional resistance. Rates of movement increased between the 4 and 8 models, but a greater increase occurred between the 8 and 12 models. A slope failure was initiated within the 16 slope model. Rates of gelifluction were dominantly influenced by increasing silt content impacting upon the distribution of segregated ice and the reduction of frictional shear strength. Increasing silt content generated high positive porewater pressures commonly in excess of hydrostatic and consequently greater amounts of pre-failure strain. A clear behavioural threshold was identified between the 10% and 20% silt soils, with far greater gelifluction in the latter than the former. Increasing clay content had a less pronounced impact upon rates of gelifluction when compared to increasing silt due to cohesion. Rates of movement increased between the 10% and 20% clay in response to lower shear strength. A sawtooth style of pore pressure response caused by water escape events within the 20% clay prevented maximum potential pressures being achieved and possibly impacted upon the overall rate of gelifluction. A successful simulation of both landsliding and slow mass wasting processes was undertaken and future applications for the technique have been outlined.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:583399 |
| Date | January 2004 |
| Creators | Smith, James Seymour |
| Publisher | Cardiff University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://orca.cf.ac.uk/54676/ |
Page generated in 0.0021 seconds