Abstract:
Past earthquakes have indicated that liquefaction-induced lateral spreading is a major cause of collapse for pile foundations. Liquefaction is a phenomenon where saturated granular soils temporarily lose strength during an earthquake. Although it is accepted that pile foundations should be designed to sustain the individual effects of lateral spreading and superstructure inertia, it is not clear if these two loads should be combined when designing pile foundations. This uncertainty has caused a great deal of confusion among designers. Some studies suggest that the two loads do interact during earthquakes (e.g., Boulanger et al. 2007). If this is the case, failure to combine the two loads will result in designs that are inadequate and unsafe. In contrary, some design guidelines suggest that lateral spreading occurs towards the end of earthquake motion and does not necessarily interact with inertial loads that occur during the motion (MCEER/ATC 2003). If this design assumption is correct, costly ground improvements can be avoided especially for piles passing through thick nonliquefiable crusts overlaying deep liquefiable soils. Design codes vary significantly on how to analyze and design piles subjected to liquefaction and lateral spreading.
This paper addresses this knowledge gap using experimental data from a 1-g shake table test on a pile embedded in gently sloped liquefiable soil underlying a nonliquefiable crust. This test was performed at the University of California San Diego (UCSD) in 2018 by Professor Ahmed Elgamal and his colleagues. The recent analysis of experimental data was sponsored by DFI project funds and co-endorsed by the DFI Seismic and Lateral Loads (S & LL) and Drilled Shafts (DS) committees. The experimental data was used to evaluate the interaction of inertial kinematic demands during seismic loading. Research results are used to discuss implications on the design of pile foundations in liquefiable soils using commonly-used tools (e.g. LPILE).