Assessing liquefaction potential of soils utilising in-situ testing.

Abstract

Liquefaction has caused significant failures and represents a significant problem for the community and geotechnical engineering designers (Pyrah et al., 1998). However, in practice, a single reliable method for assessing the liquefaction potential of soils is not well defined, particularly for aged soils. This is due mainly to the fact that most research has been based on ‘clean sand’ as the calibration to define the boundary between liquefaction and non liquefaction behaviour. Therefore, a well defined procedure for liquefaction assessment which is applicable to soils of any age is a crucial first step in reducing the risk of substructure failures and mitigating casualties resulting from earthquakes. The research presented herein is focused on investigating the capability of the cone penetration test (CPT) and flat dilatometer test (DMT) for liquefaction assessment on natural soils considering soils deposited more than 1100 years ago at Gillman, South Australia. The recommended CPT procedure from the 1996 NCEER and 1998 NCEER/NSF Workshops is employed. In addition, the age correction factor proposed by Hayati et al. (2008) is used to revise the cone resistance ratio (CRR) values obtained from the NCEER/NSF procedure. The DMT procedure is selected as another contender in this liquefaction assessment because some researchers, such as Yu et al. (1997), Sladen (1989) and Marchetti (1999), claimed that the DMT is able to capture the ageing effect of the soils. Extensive study to define the peak ground acceleration for this liquefaction assessment is conducted by using one-dimensional, site-specific ground response analysis (SHAKE91 and EERA). The most recent and significant natural earthquake motions recorded by two separate accelerogram stations are obtained and manipulated to suit the data entry format of the response analysis methods. The soil unit weight and its shear wave velocity are derived from CPT and DMT data by using several empirical correlations. The results are then applied individually to each procedure. The critical state approach for liquefaction assessment introduced by Jefferies & Been (2006) is used to verify the assessment of both the CPT and DMT procedures. The simple critical state parameter test proposed by Santamarina & Cho (2003) is undertaken on 6 soil samples taken from the study site to estimate the in-situ state parameter. Liquefaction assessment using the CPT data incorporating ageing and DMT procedures (i.e. Marchetti, 1982 and Monaco et al., 2005) are presented and a comparison between all procedures is carried out. Re-examination using critical state approach is made. In addition, the consequences of the liquefaction in terms of ground settlement are also investigated. Finally, this study shows that soil ageing increases the ability of soil to resist during the seismic loading. Furthermore, by assuming that the critical state approach represents the true conditions of the study site, the liquefaction assessment method proposed by Marchetti (1982) from DMT data provides better prediction than the others.