Relative accuracy of CPT-based liquefaction evaluation procedures: Lessons learned for the 2010-2011 Canterbury earthquake sequence

Abstract

Data from the 2010-2011 Canterbury earthquake sequence (CES) provides an unprecedented opportunity to assess and advance the current state of practice for evaluating liquefaction triggering. Towards this end, select case histories from the CES are used herein to assess the predictive capabilities of three alternative CPT-based simplified liquefaction evaluation procedures: Robertson and Wride (1998); Moss et al. (2006); and Idriss and Boulanger (2008). Additionally, the Liquefaction Potential Index (LPI) framework for predicting the severity of surficial liquefaction manifestations is also used to assess the predictive capabilities of the liquefaction evaluation procedures. Although it is not without limitations, use of the LPI framework for this purpose circumvents the need for selecting “critical” layers and their representative properties for study sites, which inherently involves subjectivity and thus has been a point of contention among researchers. It was found that while all the assessed liquefaction triggering evaluation procedures performed well for the parameter ranges of the sites analyzed, the procedure proposed by Idriss and Boulanger (2008) yielded predictions that are more consistent with field observations than the other procedures. However, use of the Idriss and Boulanger (2008) procedure in conjunction with a Christchurch-specific correlation to estimate fines content showed a decreased performance relative to using a generic fines content correlation. As a result, the fines correction for the Idriss and Boulanger (2008) procedure needs further study.