Identifying the optimal intensity measure and key factors of earthquake liquefaction-induced uplift of underground structures

Rationally identifying key factors is important for assessing the liquefaction-induced uplift of underground structures. Peak ground acceleration (PGA), which is a commonly used ground-motion intensity measure (IM) for the assessment of earthquake-liquefaction disasters, cannot describe the characteristics of earthquake motions well, and may not be the most suitable for evaluating uplift, so the identification of the optimal IM is worth studying. Therefore, taking a subway station as an example, a numerical method is used to simulate the uplift response for 74 cases in this study, and then the sensitivity analysis method is used to quantify the degree of influence of factors. Furthermore, the best IM suitable for evaluating the uplift displacement is identified from 32 candidate IMs using correlation, feasibility, and validity analysis methods. The results show that the relative density of sand (D-r), IM, thickness of the critical layer (T-s), and structural depth (Z) have much larger influences on the uplift response than other factors. Among the IMs, the Arias intensity (AI) is optimal, followed by the characteristic intensity (I-c), while PGA has shortcomings in assessing uplift response. The proposed key factors are D-r, AI, Z, T-s, structural self-weight (N), and groundwater table (D-w).