An analytical model and applications for ground surface effects for liquefaction

Many procedures for evaluation of liquefaction potential are based on field performance data. Liquefaction failure is generally considered to have occurred when pore pressures become large enough to produce sand boils, ground oscillation, lateral spreading, or other evidence at the ground surface. These features form the basis for characterizing liquefaction resistance by various approaches (cyclic stress, Arias intensity, etc.). While the cyclic stress procedure only considers the characteristics of the underlying soils, the development of sand boils also depends on other characteristics such as thickness, compressibility, permeability and intactness of the underlying soil layers. Therefore, the conditions under which sand boils form at the ground surface become an important part of the case histories database, and how the various procedures are used to evaluate liquefaction potential.; The first part of this research developed a rational method by which the initiation of liquefaction below the ground surface and the potential for sand boil formation can be evaluated. The method was designed to predict the conditions under which surface evidence of liquefaction can be expected to occur by computing a sand boil potential index, identifying the uncertainty of the prediction, and using the results to identify the conditions under which liquefaction can occur at depth without the formation of surface features caused by liquefaction.; The second part of this research presents a new method by which paleoliquefaction evidence can be used to estimate ground motion parameters. The method was a probabilistic-based approach for determining the distribution of peak acceleration that corresponds to paleoliquefaction observations. In developing a probabilistic procedure for determining any ground motion parameter from paleoliquefaction evidence, consideration must be given to the components used in traditional seismic analysis.; The remaining part of this research presents a body of work that was completing as a practical application of geotechnical earthquake engineering and liquefaction. This application proposed a probabilistic method for an early warning system for the damaging effects of strong ground shaking of the Alaskan Way Viaduct in Seattle, Washington.