Which Is a Better Proxy, Site Period or Depth, in Modelling Linear Site Response in Addition to the Average Shear-Wave Velocity?

Date: 4/25/2019

Time: 01:30 PM

Room: Pine

This study aims to identify the best site parameter to characterize empirical linear site response in addition to the conventional average shear-wave velocity. KiK-net database is utilized, and site amplification is calculated as the surface-to-borehole spectral ratios. To minimize the influence of inhomogeneous site condition at borehole stations, we include only stations of which downhole shear-wave velocity is more than 800 m/sec. Meanwhile, we rule out nonlinearity by using only ground motions of which maximum shear-strain within a soil profile below the minimum level of a possible nonlinear effect. Besides, sites with less than three records are excluded. Finally, we selected 1840 ground-motion records (six components per record) with rupture distance up to 400 km, with corner-frequency no lower than 0.12 Hz, with signal-to-noise ratio more than three and from seismic events with moment magnitude between 3.0 and 8.0.

The linear site amplification is modelled using VSz (shear-wave velocity averaged to a depth z, where z=5, 10, 20, 30 m) as a primary site proxy, and then residuals after VSz correction are modelled using an additional site parameter as a secondary proxy, i.e., T0 (site fundamental period) and Zx (site depth in meters to a layer having shear-wave velocity x, where x=0.8, 1.0, 1.5, and 2.5 km/sec). T0 was derived using the H/V technique of earthquake records. Both measured Zx from velocity profiles and inferred Zx from a regional velocity model (J-SHIS) are considered. The performances of secondary site parameters are then gauged based on the reduction in site-to-site variability due to their incorporations. Our preliminary results show that T0 is preferred as a secondary proxy over Zx to be used together with VS30 to model linear site response.

Presenting Author: Chuanbin Zhu

Authors