EGU24-7351, updated on 08 Mar 2024
https://doi.org/10.5194/egusphere-egu24-7351
EGU General Assembly 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Temporal evolution of resonance frequencies as a proxy to monitor nonlinear site response: a case study using KiK-net data in Japan

Ssu-Ting Lai1, Alessandra Schibuola2, Luis Fabian Bonilla2, and Fabrice Cotton1
Ssu-Ting Lai et al.
  • 1Helmholtz Centre Potsdam - GFZ German Research Centre for Geosciences, Potsdam 14473 , Germany
  • 2Université Gustave Eiffel, Cité Descartes, 77447 Marne-la-Vallée Cedex 2, France

The comprehension of earthquake ground motion amplification in soft sediments, influenced by the contrasting physical properties of rock and soil, faces challenges due to nonlinear site responses. Nonlinear effects, driven by distinct stress-strain behavior in soils under substantial loads, result in modified propagation velocity of the media, shifts in resonance frequencies, stronger damping, and reduced amplification. Identifying nonlinearity is particularly challenging, especially in the absence of a reference site. In addition, prevalent approaches in ground motion prediction studies often rely on empirical equations utilizing numerical computations that are difficult to validate, instead of incorporating the nonlinear effects present in the data.

In this study, we introduce a novel approach by constructing resonance curves derived from seismic waveforms recorded at the surface stations within the KiK-net network in Japan. These curves not only provide the resonance frequencies at a specific site but also serve as proxies for broadband site response, all while excluding site amplification. Our approach unveils the extent of frequency shifts in resonance frequencies, elucidating the interplay between events and sites across varying ground motion levels. Validation through borehole responses from the KiK-net network demonstrates the robustness of our methodology in characterizing the subsurface solely based on surface recordings. This contribution aims to identify both linear and nonlinear resonance frequencies using all available earthquake data at any site, which can help improve ground motion prediction studies, and more accurate seismic hazard assessments.

How to cite: Lai, S.-T., Schibuola, A., Bonilla, L. F., and Cotton, F.: Temporal evolution of resonance frequencies as a proxy to monitor nonlinear site response: a case study using KiK-net data in Japan, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7351, https://doi.org/10.5194/egusphere-egu24-7351, 2024.