1-g Shake Table Experimental Evaluation of Building Settlements Founded on Liquefiable Soils
AuthorToth, Joseph A.W.
Civil and Environmental Engineering
StatisticsView Usage Statistics
Post-disaster reconnaissance of areas affected by recent earthquakes in Japan and New Zealand has documented extensive damage to buildings with shallow foundations resulting from liquefaction-induced settlement. Current practices in predicting degree of liquefaction-induced settlement are based on semi-empirical relationships for free-field conditions and do not consider external loadings from structures. However, field observations have noted that liquefaction settlement from buildings can be considerably greater than the semi-empirical estimations. The controlling mechanisms of liquefaction settlement under load are not well understood and are currently being investigated by researchers within the Geo-seismic community. Our research is based on a series of 1-g shake table experiments using a transparent soil box to reproduce liquefaction-induced building settlements. Settlements were evaluated using a scaled model of a building foundation representative of a lightly loaded single to double story building. Experimental testing included use of manually induced shaking, implementation of an eccentric-mass shaker and use of a biaxial large scale shake table. Comprehensive parametric study was carried out to establish the effects of several parameters on free-field and building settlements such as building width, relative density, ground motion duration and thickness of liquefiable layer. Experiments included use of accelerometers, pore water pressure sensors and linear variable differential transformers (LVDT) to monitor behavior in both free-field and model building footprints during induction of liquefaction. A comprehensive parametric study was conducted evaluating the influence of key parameters. Results of this study suggest the following on liquefaction settlement behavior. Increases in foundation width showed decreases in settlement. Increases in relative density of soil also showed decreases in settlement. Increases in ground motion duration lead to increases in settlement. Increases in thickness of liquefiable layer lead to increases in settlement. Free-field settlement was predicted using two common methods in practice and compared with the settlements measured directly in our experiments. These predictions are shown to be lower than measurements observed for building foundations and also slightly over-predict settlements observed in the free-field. Results of these studies are also compared with previous centrifuge, shake table and field observations normalized for width of foundation and thickness of liquefiable layer and are generally in good agreement. Lastly, a brief discussion is presented suggesting the use of helical piles as a mitigation strategy in reducing the building settlements of structures founded over liquefiable soils.The width of the soil container used in experimentation restricted use of larger model foundation diameters. Current model diameters used in experimentation suggest that prototype foundations are more typical of isolated piers and footings rather than mat foundations when considering laws of similitude. Additionally, soil model grain size characteristics are representative of a coarse sand to fine grained gravel.