Evaluation of DMT-based liquefaction triggering curves based on field case histories

Liquefaction is commonly evaluated using in-situ tests such as the cone penetration test (CPT) and the standard penetration test (SPT). However, these tests are relatively insensitive to a number of factors that are known to influence liquefaction resistance such as aging, stress history, overconsolidation, and horizontal earth pressure. In contrast, the flat blade dilatometer test is much more sensitive to these parameters and could potentially provide liquefaction resistance evaluations which can account for these factors. Three methods are available for predicting liquefaction resistance based on the DMT horizontal stress index KD; however, their accuracy is poorly defined. To provide more direct evidence regarding the validity of the various approaches, DMT data has been collected at sites where liquefaction has and has not occurred in various earthquakes. The data set includes sites in California, Taiwan, New Zealand and Italy. In several cases, the sites were subjected to multiple earthquakes which did and didn’t induce liquefaction. The CRR vs KD curve is relatively well constrained for KD values less than about 4. Both the Tsai et al. (2009) and Robertson (2012) curves provide reasonable triggering boundaries within this range. In contrast, the Monaco et al. (2005) curveis somewhat unconservative with liquefaction points below the curve. For KD values greater than 4.0 there is insufficient data to determine which of the three triggering curves is most appropriate. Additional testing is necessary at sites with KD greater than 4.0 where CSR is higher than 0.20 to define the triggering curve in this region.