Evaluation of LPI and LSN for South Korea based on the Liquefaction Observation

Abstract

The M5.4 earthquake occurred on November 15, 2017 in Heunghae eup, Pohang city, South Korea. This earthquake was recorded as the second largest instrumented earthquake in South Korea and caused numerous damages on grounds and structures. Among the other ground deformations such as settlements, cracks, and landslides, the hundreds of liquefaction-induced sand volcanos observed near the epicenter were the major issue. Korea Institute of Geoscience and Mineral Resources (KIGAM) recorded locations of those sand volcanos. We collected the existing boring data with Standard Penetration Test (SPT) N value from the Geotechnical Information Portal System and the local government offices. We also utilized the results of the boring and SPT which were performed immediately after the Pohang earthquake. Using the SPT N values and water table information from these boring data and the peak ground acceleration values obtained from the U.S. Geological Survey (USGS) Shake map, we computed Liquefaction Potential Index (LPI) following the procedure proposed by Iwasakiet al. (1978) and Liquefaction Severity Number (LSN) based on the approach by Tokimatsu and Seed (1987). By comparing the LPI and LSN values with the locations of sand volcanos, we found that the accuracies of both the LPI and LSN are approximately 70 %. We also computed probabilities of liquefaction in functions of LPI and LSN. It turns out that the liquefaction probabilities are approximately 0.1 and 0.6 when the LPIs are 5 and 12, respectively. The liquefaction probabilities are approximately 0.1 and 0.4 when the LSNs are 30 and 70, respectively.