Three-dimensional surface wave tomography for the upper crustal velocity structure of southern Korea using seismic noise correlations

The cross-correlation technique was applied to ambient seismic noises to investigate the upper crustal velocity structure of the southern Korean Peninsula based on surface wave group velocities. We used seismograms recorded continuously during August 2005 at 91 accelerograph stations in the southern Korean Peninsula. From the correlation results of 4095 data pairs, the arrival times of Green's functions along paths between stations of pairs were measured in discrete bins of frequency bands with central frequencies of 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, and 0.45 Hz. The spatial distribution of arrival times for each frequency hand bin was inverted to obtain a two-dimensional tomographic map of Rayleigh wave group velocity. For each tomographic cell of the maps, we derived dispersion curves by extracting group velocities at center frequencies of the 7 frequency hand bins. Shear-wave velocity obtained by depth inversion of the dispersion curve at each cell produced a one-dimensional velocity profile at depths. After aligning the profiles corresponding to tomographic cells, we obtained a three-dimensional uppermost crustal velocity, model at depths down to approximately 8 kin in the southern Korean Peninsula. These results show low-velocity distributions in the Gyeongsang Basin, western Gyeonggi Massif, and the area of Jeju Island and its northeastern coast. The Okeheon Fold Belt and Yeongnam Massif have relatively high velocity distributions. The low-velocity distribution in the Gyeongsang Basin shrinks in north-south direction at depths around 4 kin and widens again with a tendency to migrate to the east as the depth is increased. However, the linear trend of low velocities in NNE-SSW direction at deeper depths (6.25-8.25 km) in the western area distinctively coincides with the western boundary of the Gyeongsang Basin. The background noise analysis in this study provided the constraining information for determination of depth variation in the shear-wave velocity distribution.