Pumice materials, which are problematic from an engineering viewpoint, are widespread in the central part of the North Island. Considering the impacts of the 2010-2011 Christchurch earthquakes, a clear understanding of their properties under earthquake loading is necessary. For example, the 1987 Edgecumbe earthquake showed evidence of localised liquefaction of sands of volcanic origin. To elucidate on this, research was undertaken to investigate whether existing empirical field-based methods to evaluate the liquefaction potential of sands, which were originally developed for hard-grained soils, are applicable to crushable pumice-rich deposits. For this purpose, two sites, one in Whakatane and another in Edgecumbe, were selected where the occurrence of liquefaction was reported following the Edgecumbe earthquake. Manifestations of soil liquefaction, such as sand boils and ejected materials, have been reported at both sites. Field tests, including cone penetration tests (CPT), shear-wave velocity profiling, and screw driving sounding (SDS) tests were performed at the sites. Then, considering estimated peak ground accelerations (PGAs) at the sites based on recorded motions and possible range of ground water table locations, liquefaction analysis was conducted at the sites using available empirical approaches. To clarify the results of the analysis, undisturbed soil samples were obtained at both sites to investigate the laboratory-derived cyclic resistance ratios and to compare with the field-estimated values. Research results clearly showed that these pumice-rich soils do not fit existing liquefaction assessment frameworks and alternate methods are necessary to characterise them.
There is a growing awareness of the need for the earthquake engineering practice to incorporate in addition to empirical approaches in evaluation of liquefaction hazards advanced methods which can more realistically represent soil behaviour during earthquakes. Currently, this implementation is hindered by a number of challenges mainly associated with the amount of data and user-experience required for such advanced methods. In this study, we present key steps of an advanced seismic effective-stress analysis procedure, which on the one hand can be fully automated and, on the other hand, requires no additional input (at least for preliminary applications) compared to simplified cone penetration test (CPT)-based liquefaction procedures. In this way, effective-stress analysis can be routinely applied for quick, yet more robust estimations of liquefaction hazards, in a similar fashion to the simplified procedures. Important insights regarding the dynamic interactions in liquefying soils and the actual system response of a deposit can be gained from such analyses, as illustrated with the application to two sites from Christchurch, New Zealand.
The Canterbury Earthquake Sequence (CES), induced extensive damage in residential buildings and led to over NZ$40 billion in total economic losses. Due to the unique insurance setting in New Zealand, up to 80% of the financial losses were insured. Over the CES, the Earthquake Commission (EQC) received more than 412,000 insurance claims for residential buildings. The 4 September 2010 earthquake is the event for which most of the claims have been lodged with more than 138,000 residential claims for this event only. This research project uses EQC claim database to develop a seismic loss prediction model for residential buildings in Christchurch. It uses machine learning to create a procedure capable of highlighting critical features that affected the most buildings loss. A future study of those features enables the generation of insights that can be used by various stakeholders, for example, to better understand the influence of a structural system on the building loss or to select appropriate risk mitigation measures. Previous to the training of the machine learning model, the claim dataset was supplemented with additional data sourced from private and open access databases giving complementary information related to the building characteristics, seismic demand, liquefaction occurrence and soil conditions. This poster presents results of a machine learning model trained on a merged dataset using residential claims from the 4 September 2010.
