日本地球惑星科学連合2021年大会

講演情報

[J] ポスター発表

セッション記号 S (固体地球科学) » S-CG 固体地球科学複合領域・一般

[S-CG50] 変動帯ダイナミクス

2021年6月3日(木) 17:15 〜 18:30 Ch.15

コンビーナ:深畑 幸俊(京都大学防災研究所)、岩森 光(東京大学・地震研究所)、大橋 聖和(山口大学大学院創成科学研究科)

17:15 〜 18:30

[SCG50-P12] Lithologic and thermal controls of seismicity in the northwestern Kii Peninsula, southern Honshu, Japan

*前田 純伶1、遠田 晋次2、松澤 暢3、大坪 誠4 (1.国立研究開発法人 防災科学技術研究所 、2.東北大学 災害科学国際研究所、3.東北大学 大学院理学研究科・理学部、4.国立研究開発法人 産業技術総合研究所 地質調査総合センター)

A spatial similarity of the distributions between inland earthquakes and the lithotectonic belts, which is considered to be controlled by the mechanical properties of the lithosphere such as rheology of minerals, has been reported by several papers (e.g., Huzita et al. 1973). Particularly, the relationship between the deeper limit of earthquakes (seismicity cut-off) and the depth to the brittle-ductile transitions in the crust has been extensively investigated (e.g., Albaric et al. 2009; Hauksson and Meier 2019). However, these studies have compared the vertical distribution of the seismicity only with the rocks consisting of the lithotectonic belts at the ground surface, but not with the subsurface geological structure. In order to understand the relationship between the earthquake occurrence and the rheological properties at depth in detail, we attempted to quantitatively interpret the horizontal variety in the vertical extent of the shallow inland seismicity related to the difference of the lithotectonic belts, based on the temperature-dependent rock friction.

The study area is the northwestern Kii Peninsula, southern Honshu, Japan. We first relocated the earthquake hypocenters in this area, based on the double-difference location method (Waldhauser and Ellsworth 2000). We then estimated the density distribution (Talwani et al. 1959) in the study area by analyzing gravity data (Geological Survey of Japan 2013) to clarify how deep the surficial lithotectonic belts vertically extend. We finally compared the horizontal and vertical relationship between the seismicity and the geological structure based on these results.

From the density analysis, we confirmed that the surface rock bodies in the tectonic belts are likely to extend from the surface to at least 7 km depth, although the lower limit of the subsurface tectonic structure can be deeper. Based on the results of the relocated hypocenters, we found that the earthquakes are concentrated mainly in the shallow part (approximately 4.5–6.0 km) within the pelitic rock belts. Within the mafic rock belts, on the other hand, the seismicity rate is low in the shallow part but high in the deep part (approximately 5.5–7.5 km).

In order to interpret the difference in the vertical extents of the seismicity depending on the lithology, we compare the frictional properties of pelitic and mafic rocks obtained by rock experiments with the earthquake depths within the lithotectonic belts composed of respective rocks in the study area. The rock experiments have shown that gouges of the pelitic and mafic (metabasalt) rocks exhibit the velocity-weakening at 250℃ to 400℃ (Den Hartog et al. 2012) and at 300℃ to 600℃ (Zhang et al. 2017), respectively. Based on the heat flow of 130 mW/m2 in the Wakayama region estimated by Matsumoto (2007) and thermal conductivity of 2.5 Wm–1K–1 (Fowler 2005; Tanaka 2009), these temperature ranges in which the velocity-weakening occurs are converted to the depth ranges of approximately 4.7–7.7 km for the pelitic belt and 5.7–11.8 km for the mafic belt. The depth ranges of high-seismicity areas within representative pelitic and mafic rock belts are consistent with these vertical ranges of the velocity-weakening found from the experiments. These good consistencies strongly suggest that the occurrence of shallow intraplate earthquakes is controlled by the temperature-dependent and lithology-dependent friction.