SEGJ14th

Presentation information

Oral presentation

Disaster Mitigation Application

Disaster mitigation

Thu. Oct 21, 2021 1:55 PM - 3:15 PM Room 2 / Oral session (Zoom 2)

Chair:Nobuo Takai

2:15 PM - 2:35 PM

[DM-06] Assessment of the Validity of the One-Dimensional Approximation in the Velocity Structure Modeling Based on the Spatial Correlation of Seismic Motions

*DAIKI TORII1, HIROAKI SATO1, SHOMA YAMAMOTO2, ATSUSHI OKAZAKI2 (1. Central Research Institute of Electric Power Industry (Japan), 2. The Kansai Electric Power Co., Inc. (Japan))

The velocity structure models used for earthquake ground motion evaluation at critical facility sites are often constructed using average 1D models based on geophysical exploration such as PS logging, seismic reflection survey, and microtremor array survey. Therefore, it is important to check the validity of the one-dimensional approximation in modeling. In this respect, it is often applied to verify the stability of the amplification characteristics using earthquake records with different directions. However, this method is not applicable when there is a bias in the epicentral distribution around the site.
In this study, we conducted a dense seismic observation at the target site. We then tried a new approach to evaluate the validity of the one-dimensional approximation in modeling at the site based on the spatial correlation of the cross-correlation coefficient between observation stations. As a result, the characteristic of the spatial correlation was getting smaller with the distance between stations and then asymptotically approached a constant value. Furthermore, this distance-dependent feature was found to have a shorter distance and a smaller constant value with a higher frequency. Finally, we evaluated the validity of the one-dimensional approximation based on the similarity of the spatial correlations in each area, which are mathematically modeled by the exponential-type autocorrelation function.

Abstract password authentication.
The password was sent to the registered participants by e-mail in advance.

Password