JpGU-AGU Joint Meeting 2020

Presentation information

[E] Poster

S (Solid Earth Sciences ) » S-SS Seismology

[S-SS04] Strong Ground Motion and Earthquake Disaster

convener:Kazuhiro Somei(Geo-Research Institute), Yasuhiro Matsumoto(Kozo Keikaku Engineering)

[SSS04-P02] Tilt and rotational motions during the strong shaking

*Toshimi Satoh1 (1.Shimizu Corporation)

Keywords:tilt, rotation, nonliear site response, dilatancy, strong motion records, consolidation settlement

Satoh (2019) showed that some velocity time histories by integrating acceleration records of vertical components have linear trends to the downward direction then become almost flat using mainly KiK-net records. Satoh (2019) defined the absolute flat value as Vnon. In this study we estimate Vnon from vertical components of K-NET, KiK-net, JMA and local government strong motion records with PGAs of larger than 300 cm/s2. The velocity time histories of horizontal records with large Vnon have linear trends during the strong shaking except for records with the largest Vnon at Chokubetsu (HKD086) of the 2003 Tokachi-oki earthquake. In other words, acceleration records of horizontal components have step functions. Therefore, we also estimate the tilt angles and the rotational angles in the horizontal plane using the step values of horizontal components of records with Vnon greater than 1 cm/s.

NIED showed by the site investigation at HKD086 after the 2003 Tokachi-oki earthquake that the foundation of the seismometer had floated up accompanied with the tilt and rotation due to the liquefaction. Since the existence of Vnon suggests the settlement of the ground, the floating up occurred after the strong shaking, that it, post-liquefaction. The velocity time histories of horizontal components at HKD086 have curved trend. This result indicates that the tilt and rotation of the ground is changing during the strong shaking. For comparison, array records observed at Port Island during the 1995 Hyogo-ken Nanbu earthquake are analyzed. The velocity time history of the vertical component at the ground surface has no Vnon. Since the liquefaction intaneously occurred and excess pore water pressure ratio kept 100 % during the strong shaking (Kawase et al.,1998), it is thought that the consolidation settlement did not occur during the strong shaking. On the other hand, duration of records at HKD086 are long and so the dissipation of the excess pore water pressure is considered to occur during the strong shaking. The velocity time history of vertical component observed at the depth of 16 m at Port Island has the linear trend to the upward direction then becomes almost flat. This phenomenon may occur by the floating up due to upward seepage flow by liquefaction.

Many records with large Vnon are observed at special soil, such as volcanic clayey soil or high organic soil and the particle orbits on the horizontal-vertical plane show tilt or rotational motions. Some records with large Vnon have been pointed out the increase of the excess pore water pressure by previous studies. The record at Kushiro observed by JMA during the 1993 Kushiro-oki earthquake has large Vnon and the predominant frequency of vertical motion is twice of horizontal components. On the other hand, the record at Kushiro observed by BRI 50m away from the JMA Kushiro with smaller ground motions than JMA Kushiro has no Vnon and the vertical motion dose not dominant at twice of the peak frequency of horizontal components. The feature that vertical motions with very large Vnon are twice of predominant frequencies of horizontal motions has been pointed out as the negative dilatancy of dry soil by experimental and analytical studies. Therefore, the difference of strong motions at two statins can be interpreted by the difference of nonlinear site responses with and without the negative dilatancy.

The directions of tilt estimated from the linear trends of horizontal velocity time histories in the near-fault regions are almost radial and the tilt starts at S-wave potions. The tilt angles are much larger than tilt angles of the crustal deformation due to source processes. We interpret from these results that the tilt and rotational motions are caused by the increase and dissipation of excess pore water pressure or the negative dilatancy.

Acknowledgements: We use NIED (K-NET, KiK-net), JMA, local government, BRI, and PARI strong motion records.