UC Berkeley

The 2010-2011 Canterbury earthquake sequence (CES) devasted Christchurch, New Zealand. Liquefied soil were ejected onto the ground surface, damaging more than 15,000 houses beyond economic repair. Thick, clean sand deposits were largely responsible for the observed ejecta-induced damage. However, strongly shaken silty sites did not exhibit evidence of liquefaction despite the expectation that they would, based on the existing empirical liquefaction triggering methods. At silty soil sites, liquefaction triggering methods indicate expensive ground improvements and robust foundations are necessary, but field evidence during the CES indicate they are not required. This research addresses ejecta-induced settlement at all sites and the potential for liquefaction manifestation at silty soil sites.Comprehensive assessment and documentation of liquefaction-induced land damage resulted in an unparalleled opportunity to study the effects of liquefaction. In this study, 235 case histories that document the occurrence and quantity of ejecta and its effects on infrastructure are compiled. Fifty-eight sites in the database are each analyzed for the four main earthquakes of the CES, and three additional sites are analyzed for the first event of the CES only because it did not induce lateral spreading. Direct measurements of ejecta were not conducted for the CES; hence, the ejecta-induced settlement values are estimated using LiDAR-based and photographic-based approaches. The information related to ground conditions and seismic demand leading to differing quantities of ejecta-induced settlement during the CES are also described. This unique database of detailed ejecta case histories was then examined for general trends associated with the severity of liquefaction ejecta-induced settlement and to evaluate the efficacy of some of the liquefaction-induced damage indices. Most sites in the database can be characterized by thick, clean sand deposits, which frequently underwent severe-to-extreme ejecta-induced settlement, especially as the equivalent Mw = 6.1 PGA exceeded 0.40 g. The severe-to-extreme ejecta-induced settlement tends to be systematically underestimated by current state-of-practice liquefaction-induced damage indices, especially if they do not incorporate the post-shaking hydraulic mechanisms in their formulation. The systematic underestimation of liquefaction ejecta at thick, clean sand sites, and, similarly, the systematic overestimation of liquefaction ejecta at stratified silty soil sites suggests the importance of the seismic soil system response, which is not considered in the simplified liquefaction triggering methods. Additionally, the severely damaged land by the Feb 2011 liquefaction ejecta formed cracks and defects in the non-liquefiable crust which liquefied soil at depth could exploit to form ejecta at the ground surface during the Jun 2011 earthquake. Therefore, the liquefaction ejecta case histories provide a sound basis for the investigation of the occurrence and effects of ejecta and the development of a procedure to estimate the quantity of ejecta in an earthquake. The first cyclic simple shear tests on high-quality retrieved specimens of Christchurch silty and sandy soil were performed to evaluate their liquefaction potential. The soil’s undrained cyclic stress-strain responses indicate cyclic mobility. The differences in the cyclic responses of the soil with different amounts of non-plastic silt (2%-99% fines) are subtle. The soil stiffness was significantly lower in post-cyclic tests than in monotonic tests due to the large differences in the initial effective stresses between the two types of tests, loss of strength, age, and fabric effects. The field-adjusted laboratory-based cyclic resistances agree with CPT-based cyclic resistances from simplified liquefaction triggering procedures. Both assessments indicate the silty soil deposits generated high excess pore water pressures and liquefied during the Christchurch earthquake even though surface manifestations of liquefaction were not observed at the considered sites. The absence of liquefaction manifestation at the stratified silty soil sites highlights the importance of their system response. An alternative method to the Japanese Standard JIS A 1224:2009 Test Method for evaluating the dry minimum and maximum densities of soil is proposed for those cases when a test specimen does not provide enough material to use the standard-size mold. The alternative test method was applied to soil from 42 small simple shear test specimens of retrieved high-quality soil samples. Soil from test specimens of the same material were subsequently mixed to produce eight soil composites for which their minimum and maximum densities could be determined using both the standard and alternative methods. The minimum and maximum void ratios determined by both methods for the composite soil are in good agreement. Moreover, trends in the dataset are consistent with those from previous studies on sand and non-plastic fine-grained soil. The alternative and standard test methods can be used to estimate the maximum and minimum void ratios of non-plastic silty soil with up to 70% fines so that relative density can be used to describe the state of the soil.