The 2017 Pohang earthquake [the second largest local magnitude (M-L) of 5.4 since 1978] caused significant damage: numerous sand boils and a few building settlements were observed in rice paddies and residential areas, respectively, representing unprecedented case histories of earthquake-triggered liquefaction and cyclic softening. This study evaluated liquefaction triggering and cyclic softening potentials using three in situ tests [standard penetration test (SPT), cone penetration test (CPT), and downhole (DH) test for shear wave velocity (V-S)] and laboratory tests (grain size and soil indices) for the observed sand boils and building settlements. We selected six sites, four of which had sand boils (Sites 1, 2, 3, and 4), and two of which had experienced building settlements that may have resulted from cyclic softening (Sites 5 and 6). The SPT, CPT, and V-S adequately assessed liquefaction triggering [i.e., factor of safety (FS)2 at all depths. The site-specific cyclic stress ratio through the maximum shear stress ratio computed from site response analysis appropriately evaluated the liquefaction triggering and cyclic softening at the considered sites. The results of the soil index test are consistent with the liquefaction and cyclic softening susceptibility criteria for fine-grained soils. We publicly provide the field and laboratory measurements in this study to enrich case history data on liquefaction and cyclic softening induced by intermediate-size earthquakes (e.g., a moment magnitude, M<6), which might significantly contribute to geotechnical earthquake engineering and engineering geoscience communities

Soil liquefaction by earthquake results in significant soil deformation and can cause substantial damage to infrastructure. Accordingly, there is a critical need to analyze earthquake case histories, and such analysis should focus on elucidating ground motion characteristics and field measurement data associated with instances of soil. The most straightforward approach to identifying liquefaction triggering involves examining pore water pressure records, which is challenging in many susceptible areas due to the absence of installed pore water pressure transducers. Therefore, evaluating liquefaction commonly relies on surficial evidence, such as sand boil or lateral spreading. However, this evaluation can miss the liquefaction triggered case without such surficial evidence. This study modeled the 2017 Pohang liquefaction, a first-time occurrence in South Korea, through a centrifuge test replicating the liquefied site based on field investigations, including borehole tests and recorded earthquake motions. We comprehensively assessed liquefaction using the ratio of excess pore water pressure alongside analyses of acceleration time histories, shear stress-strain hysteresis, and time-frequency histories. These results were compared with a conventional method that overlooked pore water pressure, leading to overestimation. Furthermore, using a simplified method, we compared liquefaction triggering evaluation results from the centrifuge cone penetration test and on-site standard penetration test. This, along with the factor of safety, substantiated the validity of the centrifuge results.

The 2017 Pohang earthquake, with a moment magnitude (M) of 5.5, caused severe building damage and widespread liquefaction. In this study, we evaluate the applicability of ground response and liquefaction triggering analyses for the Pohang earthquake using deep shear wave velocity (VS) profiles. The VS profiles are obtained at Handong University and the Songdo Pine Forest by inverting the Rayleigh wave dispersion curves based on microtremor array measurements (MAM) and multi-channel analysis of surface waves (MASW). In onedimensional effective stress analyses for the two sites, we consider the uncertainty of the nonlinear soil properties for three cases and use 118 rock outcrop motions. At Handong University, the spectral accelerations of surface ground motions are larger than those of the current Korean design spectra with a return period of 500 years at the natural period of the damaged buildings. At the Songdo Pine Forest, for the Case 2, numerous ground motions result in the maximum pore water pressure ratio of 1 (i.e., liquefaction occurrence). Furthermore, we calculate the liquefaction potential index (LPI) values using the VS-based simplified method. To compute the cyclic stress ratio for depths, we utilize the peak ground accelerations estimated by ground response analyses and estimated by stress reduction factor (rd), respectively. The LPI values, based on the ground response analyses, range from 0 to 4, indicating minor or no damage, while the LPI value using the rd is zero. The results of the ground response and liquefaction triggering analyses are similar to the actual damage cases.