16:06 〜 16:24
[SIT19-09] Preliminary Results on Pseudo-3D Seismic Depth Imaging of the 2024 Noto (M7.6) Earthquake Rupture Area
キーワード：Noto Earthquake, Depth Imaging, Marine Survey, 3D imaging
Prompted by the M7.6 earthquake near the Noto Peninsula on January 1, 2024, a 2D multi-channel seismic (MCS) survey was conducted in March 2024 to investigate the tsunamigenic rupture zone. Initial processing of the acquired MCS data reveals the following key findings: (1) a highly deformed zone, approximately 2–3 km wide and 30 km long, characterized by active reverse and strike-slip faults near F43 of the MLIT fault model, with notable tsunamigenic potential; and (2) an active reverse fault near the northeastern part of F42 in the MLIT fault model, exhibiting significant uplift and deformation in its hanging wall, including slope failure structures likely resulting from repeated fault activity. To further illuminate the complex geometry of the rupture zone, the Atmosphere and Ocean Research Institute (AORI) of the University of Tokyo planned a high resolution pseudo-3D MCS survey to generate a depth image over a 10 km × 10 km area. During the first run of the MCS marine survey in January 2025, approximately 5% of the total planned lines (9 lines) were recorded during the R/V Hakuho-maru cruise using a 1350 m long streamer cable with 216 active channels at 6.25 m group spacings and two GI guns with a total volume of 710 cubic inches.
Here, we present preliminary results on seismic depth imaging applied to this partial dataset. We implemented an advanced depth imaging workflow on the recorded data set to compute an optimal interval velocity model consistent with the acoustic properties of the complex geological structures in the rupture area. The workflow consists of an iterative grid-based tomography process designed to simultaneously improve extracted horizons identified by automatic pickers and nullify the residual moveout (RMO) in the depth-migrated gathers. Structural attributes are combined for the automatic picking of continuous horizons, identified through the creation of pencils. The generated pencils undergo quality control based on a list of moveout groups. Metrics such as autopick semblance, residual moveout, parameter semblance, and stack are analyzed to distinguish primary reflections from multiples, as well as true horizons from isolated spikes or noise. Based on this analysis, various moveout group masks are integrated to define the selected horizon segments, which are then used as input for the grid-based tomography step. Finally, 4 to 5 rounds of grid-based tomography refine the P-wave velocity model, improving the alignment of seismic horizons and minimizing RMO in the calculated semblances of migrated seismic gathers. The results of this study provide crucial insights into the complex subsurface geometry of the tsunamigenic fault associated with the 2024 M7.6 earthquake, enhancing our understanding of fault structures and their potential implications for seismic hazards and tsunami generation.
Here, we present preliminary results on seismic depth imaging applied to this partial dataset. We implemented an advanced depth imaging workflow on the recorded data set to compute an optimal interval velocity model consistent with the acoustic properties of the complex geological structures in the rupture area. The workflow consists of an iterative grid-based tomography process designed to simultaneously improve extracted horizons identified by automatic pickers and nullify the residual moveout (RMO) in the depth-migrated gathers. Structural attributes are combined for the automatic picking of continuous horizons, identified through the creation of pencils. The generated pencils undergo quality control based on a list of moveout groups. Metrics such as autopick semblance, residual moveout, parameter semblance, and stack are analyzed to distinguish primary reflections from multiples, as well as true horizons from isolated spikes or noise. Based on this analysis, various moveout group masks are integrated to define the selected horizon segments, which are then used as input for the grid-based tomography step. Finally, 4 to 5 rounds of grid-based tomography refine the P-wave velocity model, improving the alignment of seismic horizons and minimizing RMO in the calculated semblances of migrated seismic gathers. The results of this study provide crucial insights into the complex subsurface geometry of the tsunamigenic fault associated with the 2024 M7.6 earthquake, enhancing our understanding of fault structures and their potential implications for seismic hazards and tsunami generation.