今週のうなぎセミナーについてお知らせいたします。

Here is information of the Unagi-seminar(June, 19th).

************** Seminar on Seismology IV A, C /地震学ゼミナールIV A, C (Unagi Seminar) **************

科目：地震学ゼミナールIV A, C / Seminar on Seismology IV A, C（修士・博士）
日時：2025年 6月 19日 (木) 13:30~
場所：京都大学 防災研究所 本館E-232D

Date and Time：2025-06-19(Thursday), 13:30~
Place：Uji Campus Main Building E232D

---------- ---------- ---------- ----------

Speaker 1（発表者）:　船曵 祐輝 (Yuki FUNABIKI)
Title（題目）:
An Attempt of Detecting Stopping Phase from Distributed Acoustic Sensing

Abstract（要旨）:
How does an earthquake occur? What is the difference between a rupture that grows into a large event and one that stops as a small event? These are eternal themes of seismology, and understanding the rupture process of small to moderate earthquakes is essential to face this challenge. However, we must deal with high-frequency waves, which is difficult to utilize for waveform inversion, to understand the rupture process of such events, and studies of the rupture processes of small to moderate earthquakes have not made significant progress.
Although it is difficult to describe the rupture process of such events as a continuous surface, it is still possible to represent them at points. Chang & Ide (2021) approached the rupture process of M3–4 earthquakes by representing both the hypocenter and the centroid as points. Their analysis revealed two key findings: (1) Regardless of the final magnitude, hypocenters tend to concentrate in specific areas, or “hotspots.” (2) Even for earthquakes of similar size that rupture similar patches (i.e., with close centroids), their hypocenters do not necessarily coincide. These findings demonstrate the hierarchical nature of earthquake rupture based on observational evidence.
Building on their findings, a question arises: Why do earthquakes that initiate from the same hypocenter evolve into events with different magnitudes? What accounts for this difference? One possible approach to address this question is to capture the stopping phase, which marks the termination of rupture. Theoretical studies suggest that under a crack-like source model, high-frequency waves are radiated both at the onset and at the termination of rupture, and the high-frequency wave radiated at the end of rupture is referred to as the stopping phase. The existence of stopping phases has also been identified in observational records (e.g., Imanishi & Takeo, 2002). However, their detection remains challenging because high-frequency signals are easily attenuated by structural heterogeneities.
Distributed Acoustic Sensing, (i.e., DAS) is an emerging technology that has great potential in seismology. DAS enables ultra-dense seismic observations by using optical fiber cables themselves as seismic sensors, which is impossible with conventional instrumentation. Leveraging this advantage, Funabiki & Miyazawa (2025) estimated focal mechanisms from the high-frequency components of DAS-recorded waveforms. This approach offers the potential to capture stopping phases, which are often unclear at a single seismic station, and to locate their positions simultaneously with the hypocenter and centroid. Such capabilities could significantly improve our understanding of the rupture processes of small to moderate earthquakes.
As a first step toward these objectives, we applied PhaseNet-DAS (Zhu et al., 2023) to DAS recordings from the Noto region. Based on the detected phases, we performed cross-correlation between near channels to refine arrival times and redetermined their hypocenters. After that, we detected stopping phases from ordinary seismograms following the procedure of Imanishi et al. (2002). In this presentation, we discuss the results of these analyses.

* * * * * * * * * * * * * *

Speaker 2（発表者）:　小松 理子(Riko KOMATSU)
Title（題目）:
陸上GNSSデータと海域GNSS-Aデータを用いた日向灘におけるプレート間固着の推定

Abstract（要旨）:
日向灘は、南海トラフ巨大地震の震源域西端に位置し、過去には単独でM7〜8級の地震も発生していることから、将来の地震発生ポテンシャルを評価する上で、この地域のプレート間固着の推定は重要である。日向灘におけるプレート間固着に関する研究（Wallace et al., 2009など）は行われてきたが、当時はGNSS観測点が限られ、上盤プレートの内部変形のモデル化に任意性があるなどの課題があった。本研究では、1996年から2024年のGNSSデータ（大学設置点を含む）と海域のGNSS-Aデータを用い、地殻変動の様子からプレート間固着の空間分布とその時間変化を明らかにすることを目指す。
本発表では、陸域を九州南部とそれ以外に分割した単純なブロック形状を仮定したブロック断層モデル(McCaffrey, 2009)を用いて、2020年4月から2024年3月までのGNSSデータ、および2020年8月から2024年7月までの海域GNSS-Aデータから、４年間の平均的なプレート間固着状況とブロック運動の同時推定を行った。陸域のブロック境界は、南九州せん断帯に沿って設定をした。推定結果から、2024年8月に発生した日向灘における地震の震源域には、固着が見られない結果となった。そこで時系列データより、より西向きの変動の大きい、2003年5月から2005年4月までの期間で同様に推定を行った結果、2024年8月に発生した日向灘における地震の震源域には、固着が見られた。
さらに、解析期間を広げるために、L1トレンドフィルター(Kim et al., 2009)を用いて、観測点の変位速度の変換点を抽出し、客観的に分割した期間ごとの観測点変位速度の推定を行った。このような期間分割方法についても発表する。