Civil Engineering Departmental Seminar - Stress- and Energy-based methods for evaluating seismic soil liquefaction:  past, present, and future

Stress- and Energy-based methods for evaluating seismic soil liquefaction:  past, present, and future

This talk reviews recent work on cyclic stress and energy-based methodologies for estimating the demand-capacity state for initial soil liquefaction, the rise in pore water pressure, and the timing of soil liquefaction.  Field methods, such as Standard Penetration Test (SPT), Cone Penetration Test (CPT), and shear-wave velocity (Vs), provide a means to determine soil's seismic resistance to liquefaction through routine field procedures, and, in the case of Vs, a fundamental soil property. This talk will present the results of a three-decade-long international project to gather site data and develop probabilistic correlations for seismic soil liquefaction triggering. The first correlations, published in 2004, 2006, and 2013, analyzed hundreds of sites from the literature and our acquisition of 310 Vs tests at sites in China, Japan, Taiwan, Greece, and the United States. We are updating these correlations and have expanded the data set to approximately 650 sites, mainly through new testing of the M9.0 2011 Tohoku Earthquake, Japan, and the 2011 Christchurch and 2010 Darfield Earthquakes, New Zealand.  Bayesian regression and structural reliability methods facilitate a probabilistic treatment of the Vs catalog for performance-based engineering applications. Analysis of the uncertainties of the variables comprising both the seismic demand and the soil capacity is integral to the study and helps reduce overall model error.

Unlike stress-based methods, energy-based approaches use a scalar, cumulative parameter, making them well-suited for modeling pore pressure buildup and liquefaction onset. The rise in pore pressure and the onset of liquefaction can be estimated by summing the cumulative dissipated hysteretic energy normalized by the effective stress, which correlates strongly with the pore-pressure ratio r_u. Energy dissipated beyond the liquefaction boundary maintains r_u = 1.0. A key benefit of energy-based methods is improved estimation of liquefaction timing. These approaches aim to link total energy demand (Arias Intensity) to partially absorbed work ( dissipated energy).