Seismic initiation of submarine slope failures using physical modelling in a geotechnical centrifuge

• Main Author: Coulter, Stephen Edwin, 1977-
• Other Authors: Memorial University of Newfoundland. Faculty of Engineering and Applied Science
• Format: Thesis
• Language: English
• Published: 2008
• Subjects: Induced seismicity > Simulation methods; Slopes (Soil mechanics) > Stability; Submarine topography; Fraser River; Newfoundland studies; University of Newfoundland
• Online Access: http://collections.mun.ca/cdm/ref/collection/theses4/id/82568

Thesis (M.Eng.)--Memorial University of Newfoundland, 2008. Engineering and Applied Science Includes bibliographical references (leaves 388-403) The stability of offshore slopes is a major consideration in the development of both offshore and nearshore areas. The consequences of such slope failures can include the destruction of adjacent facilities, as well as the production of dangerous tsunamis. This phenomenon poses a unique and evident threat to human populations as well as valuable infrastructure. Most of these types of failures have occurred in prehistoric times and for the most part the initiation mechanisms behind them remain unobserved. One such triggering mechanism of these failures is seismic movement or in more common terms, an earthquake. -- This thesis presents a research program into the physical centrifuge modelling of the seismic initiation of submarine slope failures. The effects of impermeable layer presence, earthquake magnitude, and a phenomenon known as "seismic strengthening" are investigated. Properly scaled centrifuge modelling has been proven to be a useful tool in observing geotechnical engineering situations that would otherwise be costly or impractical to investigate due to financial and time constraints. -- A series of five centrifuge tests were performed on idealized slope geometries at a scale of 1:70. These tests were designed for ease of comparison with finite element analyses, with some associated compromises compared to field conditions. Generally, the test geometries consisted of a 2:1 slope constructed using Fraser River sand in a strongbox with a rectangular inner plan area. Models were either tested with the presence of a buried and draped silt layer, an inclined silt layer featuring an approximate 5.5:1 profile, or with no silt layer present at all. Much of the equipment and procedures required for this testing at the C-CORE Centrifuge Centre were developed by the author based on the experiences from other centrifuge centres. Models were air pluviated to obtain a target relative density of 40% and then saturated with a viscous pore fluid to achieve similitude of both static and dynamic scaling laws. Following construction procedure, models were tested in the Earthquake Simulator that is situated upon the C-CORE centrifuge and spun to a test level of 70 g. The response of the models to various earthquake loadings was observed with a high-speed data acquisition system. These responses primarily consisted of short-term and long-term data collected from installed accelerometers and pore pressure transducers, as well as other instruments used to observe the vertical and horizontal displacements of the model. -- Analysis consisted of examining the test data, as well as comparing analogous model tests to determine the effects mentioned above. The presence of a relatively impermeable silt layer in an appropriate orientation was found to increase the possibility of instability. A dilative response, characterized by observed upslope acceleration spikes coupled with negative spikes in pore pressure was observed in models that featured a silt layer as well as in the model that did not feature a silt layer. Seismic strengthening, through the process of densification due to the application of small seismic movements, was also observed to occur. Increased dilative response also occurred with increased earthquake magnitude, but the increase in failure resistance caused by this dilative response was overcome by the delayed dissipation of generated excess pore pressure underneath the relatively impermeable silt layer. Slope failure was characterized by long-term horizontal and vertical slope movements that continued, and sometimes recommenced, after the cessation of earthquake shaking, short-term slope face surface heave, and the evidence of silt layer movement in post-test observations.