[S-CG45] Science of slow-to-fast earthquakes

[SCG45-07] S-netデータを用いた高分解能S波速度構造モデルに基づく日本海溝沿いのアスペリティマッピング

*高木 涼太¹、西田 究² (1.東北大学大学院理学研究科附属地震・噴火予知研究観測センター、2.東京大学地震研究所)

キーワード：アスペリティ、日本海溝、巨大地震、微動、S波速度構造、S-net

Asperity is a frictionally locked potion on a fault plane. Accumulated shear stress at asperities is released during earthquakes, and the asperity distribution controls earthquake rupture. The knowledge of the fault asperity distributions relies on the slip inversions of past earthquakes, but the time period of modern observations is limited as compared with the recurrence intervals of large earthquakes. Although geodetic observations also provide the spatial distribution of the current locking state, the result is a snapshot during only a few decades, and the spatial resolution is generally limited due to sparse observations in ocean areas. Here, we demonstrate a new approach to map fault asperities using a high-resolution S-wave velocity structure model. With technical developments in ambient noise surface wave tomography, we found high S-wave velocity anomalies around the plate boundary along the Japan Trench using the S-net ocean-bottom seismic network data. The high-velocity anomalies are located where the shear stress was released by the 2011 Mw 9.1 Tohoku earthquake. They also fill the spatial gaps of tremors that distribute along the Japan Trench, and low-velocity anomalies along the plate boundary are well correlated with the tremor distribution. We model elastic deformation and demonstrate that assigned stress drop within the high-velocity anomalies reproduces the slip distribution of the 2011 Tohoku earthquake with high accuracy, including the >50 m slip near the trench and its moment magnitude. The results suggest that the high-velocity anomalies represent the asperities along the plate boundary. The complementary distributions of the asperities with high S-wave velocity and tremor with low S-wave velocity indicate the distribution of pore fluid pressure is the primal factor controlling the megathrust slip behavior. This study shows that high-resolution S-wave velocity structure imaging can map fault asperities and predict the slip distributions of megathrust earthquakes quantitatively.