Assessment of Aging Correction Factors for Liquefaction Resistance at Sites of Recurrent Liquefaction

Increases in the strength and stiffness of sands with time have been widely reported. These temporal gains, or “aging effects”, are discernible from both in-situ penetration resistance and liquefaction resistance (CRR), where the rate of increase in CRR exceeds that inferred from in-situ penetration data. As such, it has been recognized that aging effects may be resolved into gains measurable by in-situ penetration tests (i.e. SPT, CPT) and gains in liquefaction resistance associated with small-strain soil fabric phenomenon that are not detected by large-strain penetration tests. Because empirical liquefaction triggering curves (SPT- and CPT-based) are developed exclusively from post-liquefaction penetration resistance of Holocene-aged sands, it has been proposed that liquefaction evaluations of older deposits should correct for the effects of aging on in-situ penetration as well as liquefaction resistance. With respect to the latter, several relations have been proposed in the literature. Although principally developed to assess liquefaction resistance of Pleistocene-aged sands, regression of these relations produces a reference age (i.e. correction factor of 1) such that liquefaction resistance is stated to decrease for deposits younger than the reference age. Based on these relations, the liquefaction resistance of a previously liquefied deposit is assumed to be reduced during subsequent earthquakes if the reference age is greater than the time between prior liquefaction and subsequent seismic loading. As such, for sites of repeat liquefaction events closely spaced in time, the susceptibility of liquefaction triggering and severity of liquefaction manifestation may be greater than expected. Observations of the site-to-source distance of the most distal liquefaction site being greater for aftershock events than for equivalent-magnitude mainshocks may give credence to the validity of such corrections. However, field assessments of aging correction factors are very limited and their accuracy is thus highly uncertain. Accordingly, to evaluate whether aging corrections are applicable when extrapolated to very short ages (i.e. days to months), the study presented herein evaluates sites of recurrent liquefaction during the 2010-2011 Canterbury, New Zealand earthquake sequence. Aging relations from the literature are applied within the framework of a liquefaction potential index (LPI) assessment to determine whether their use is consistent with the observed severity of liquefaction manifestation. Based on the results of this analysis, implications to liquefaction hazard assessment are discussed, and recommendations for the use of aging correction factors are given.