No abstract

Ground-motion prediction models (GMPMs) have been developed to estimate seismic intensity considering earthquake magnitude, source-to-site distance, site condition, and so on. This study proposes GMPMs to predict 5% damped pseudospectral acceleration (PSA) for 27 periods ranging from 0.01 to 10 s in Korea, based on three machine-learning techniques (i.e., artificial neural network [ANN], random forest [RF], and gradient boosting [GB]). We use 1189 ground motions recorded at 50 surface stations during the 77 earthquakes with a local magnitude (ML) greater than 3.0, including the Gyeongju and Pohang earthquakes with ML of 5.8 and 5.4, respectively. We compare the performances of the three machine-learning-based models and the classical regression-based model in terms of the coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), standard deviation of residuals, and between-event and within-event residuals. The GB-based model shows the best performance. In addition, we analyze the working process of the GB-based model using variable importance and partial dependence (PD) plots. Among the five independent variables (ML, epicentral distance [Repi], average shear-wave velocity of the upper 30 m [VS30], focal depth, and slope angle) used in this study, ML and Repi are the most influential variables and show strong correlations with PSAs. We apply the GB-based model to three recent earthquakes larger than ML 3.0, and the model accurately predicts the PSAs at various stations. We also generate maps of estimated PSA (PSAest) values for the four periods (T = 0.01, 0.1, 1, and 3 s) for the scenario earthquake with an ML of 5.0. We provide a method for training the GB-based model using the Python library, which can enhance the ground-motion prediction not only in Korea but also worldwide, and an executable version of the validated GB-based model.

No abstract

The 2017 ML 5.4 Pohang, South Korea earthquake, which caused severe damage, raised attention to the importance of the characterization of the Pohang Basin structure and the identification of active faults. We utilize 366 seismograms recorded at 102 densely deployed seismic stations during 18 microearthquakes with ML ranging from −1.11 to 1.70. We perform the P-wave seismogram method, which has been validated in various regions using seismograms for moderate earthquakes, to estimate the time-averaged shear-wave velocity from the surface to a depth of 30 m ( VS30) in and around the basin. We perform multichannel analyses of surface waves at the 32 sites to obtain shear-wave velocity ( VS) profiles, and additionally acquire the 22 existing VS profiles within the study area. We observe that velocity seismograms predict VS30 more accurately than acceleration and displacement seismograms do. The estimated VS30 values are in good agreement with the measured VS30 values. Approximately 93.3% of the measured and estimated VS30 values are located between the ±100% difference lines. We propose a VS30 map for the vicinity of the Pohang Basin by geospatially interpolating both the measured and estimated VS30 values, which is consistent with the local topography and geology. The proposed map exhibits areas with strong VS30 contrast within the Quaternary and Tertiary sediments, which might be attributed to the existence of faults.

No abstract