Probabilistic Analysis of Local Liquefaction Potential Based on Spatial Variability of SPT Data

Abstract

In this paper, the spatial distribution of liquefaction potential is estimated using in-situ data from the Standard Penetration Test (SPT). For this purpose, a case study of a liquefiable soil at the Azad University of Qeshm is selected in the numerical modeling. After conducting the site investigation and determining SPT results at four boreholes, two distinct modeling approaches are implemented to evaluate the Liquefaction Potential Index (LPI) at the considered site; In the first method, the conditional random field for SPT data is generated in a layer-by-layer strategy and then, the LPI is obtained using a SPT-based empirical relations at each elemental column. On the other hand, in the second method, the LPI is first determined at each borehole location and then, this parameter is adopted as a stochastic variable in the construction of surficial conditional random field. It can be concluded that both approaches are able to capture the varying severity levels of liquefaction at most locations across the area of study. However, the comparison shows that using the first approach results in a more fluctuated LPI results with almost the same extremum values.