9:00 AM - 9:30 AM
[MIS24-01] Reexamination of the Fault Model for the 1707 Hōei Earthquake
★Invited Papers
Keywords:Nankai trough, The 1707 Hoei earthwuake, fault model, historical record
The 1707 Hoei earthquake is considered one of the largest Nankai Trough megathrust earthquakes in Japanease history. Previously, the authors have reorganized tsunami traces and crustal deformation distributions from the Ansei-Tokai, Ansei-Nankai, and Showa-Tonankai earthquakes and reevaluated their source model (e.g., Imai et al., 2021). To clarify the diversity of Nankai Trough megathrust earthquakes, it is necessary to reexamine the fault model for the Hoei earthquake. For this purpose, we have systematically reexamined a vast collection of historical records and reassessed crustal deformation and tsunami traces based on these records (Imai et al., 2023). This study aims to reconsider the source fault model of the Hoei earthquake by incorporating a three-dimensional plate structure into the arrangement of small fault segments and by analyzing the distribution of crustal deformation and tsunami heights.
For tsunami trace values, we utilized both existing tsunami trace data stored in the Tsunami Trace Database (Tohoku University, Nuclear Regulation Authority) and newly identified tsunami trace values obtained through a reexamination of historical records. The crustal deformation values were evaluated based on descriptions in earthquake historical archives (Imai et al., 2023). The source fault model was constructed with reference to the three-dimensional structural model of the Nankai Trough subduction zone (Nakanishi et al., 2018), assuming 30 small fault segments in total: six in the Tokai source region, ten in the Tonankai source region, and fourteen in the Nankai source region. The crustal deformation due to slip on each small fault was calculated using Okada's method (Okada, 1985), while the tsunami Green's function was computed based on the linear long-wave theory with a spatial grid spacing of 150 m and a time interval of 0.2 s.
For crustal deformation analysis, 43 observation points were used. In terms of tsunami trace data, a total of 298 data points from previous studies were available. However, to account for spatial grid limitations and the need for detailed topographic reconstructions, tsunami trace points from enclosed sea areas were excluded, and overlapping records from the same settlements were consolidated, resulting in a total of 92 points being utilized.
The slip distribution of each small fault segment was estimated using Simulated Annealing (SA) (Kirkpatrick et al., 1983) to optimize the reproducibility index VRS (Imai et al., 2020), aiming for a value close to 1. To account for uncertainties, a uniform random error of approximately 10% was introduced, and an ensemble mean of 1,000 trials was used to evaluate the slip distribution. Figure 1 shows the estimated slip distribution of the source fault. The VRS value was found to be 0.76 ± 0.01. Crustal deformation was generally well reproduced, and most tsunami traces were also accurately reconstructed. The estimated moment magnitude of the earthquake was Mw 8.7 ± 0.1, which is comparable to or slightly larger than previous evaluations. In the Tokai source region, an average slip of 7 m was required to explain the subsidence in Shimizu, Shizuoka Prefecture, and the uplift at Yokosuka Castle. In contrast, the Tonankai source region did not exhibit significant slip compared to the Ansei-Tokai earthquake (Imai et al., 2021). However, significant localized slip was required to explain the uplift observed around Kumano, Mie Prefecture. In the Nankai source region, over 10 m of slip was necessary offshore to account for tsunami heights in Tosa Bay. However, the model failed to reproduce the maximum tsunami height of 25 m recorded at Usa Shoryuji Temple. This discrepancy suggests that additional secondary tsunami sources may need to be considered.
Acknowledgments: This study was supported by the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) "Nankai Trough Earthquake Research Project for Disaster Mitigation" from fiscal year R2-6.
For tsunami trace values, we utilized both existing tsunami trace data stored in the Tsunami Trace Database (Tohoku University, Nuclear Regulation Authority) and newly identified tsunami trace values obtained through a reexamination of historical records. The crustal deformation values were evaluated based on descriptions in earthquake historical archives (Imai et al., 2023). The source fault model was constructed with reference to the three-dimensional structural model of the Nankai Trough subduction zone (Nakanishi et al., 2018), assuming 30 small fault segments in total: six in the Tokai source region, ten in the Tonankai source region, and fourteen in the Nankai source region. The crustal deformation due to slip on each small fault was calculated using Okada's method (Okada, 1985), while the tsunami Green's function was computed based on the linear long-wave theory with a spatial grid spacing of 150 m and a time interval of 0.2 s.
For crustal deformation analysis, 43 observation points were used. In terms of tsunami trace data, a total of 298 data points from previous studies were available. However, to account for spatial grid limitations and the need for detailed topographic reconstructions, tsunami trace points from enclosed sea areas were excluded, and overlapping records from the same settlements were consolidated, resulting in a total of 92 points being utilized.
The slip distribution of each small fault segment was estimated using Simulated Annealing (SA) (Kirkpatrick et al., 1983) to optimize the reproducibility index VRS (Imai et al., 2020), aiming for a value close to 1. To account for uncertainties, a uniform random error of approximately 10% was introduced, and an ensemble mean of 1,000 trials was used to evaluate the slip distribution. Figure 1 shows the estimated slip distribution of the source fault. The VRS value was found to be 0.76 ± 0.01. Crustal deformation was generally well reproduced, and most tsunami traces were also accurately reconstructed. The estimated moment magnitude of the earthquake was Mw 8.7 ± 0.1, which is comparable to or slightly larger than previous evaluations. In the Tokai source region, an average slip of 7 m was required to explain the subsidence in Shimizu, Shizuoka Prefecture, and the uplift at Yokosuka Castle. In contrast, the Tonankai source region did not exhibit significant slip compared to the Ansei-Tokai earthquake (Imai et al., 2021). However, significant localized slip was required to explain the uplift observed around Kumano, Mie Prefecture. In the Nankai source region, over 10 m of slip was necessary offshore to account for tsunami heights in Tosa Bay. However, the model failed to reproduce the maximum tsunami height of 25 m recorded at Usa Shoryuji Temple. This discrepancy suggests that additional secondary tsunami sources may need to be considered.
Acknowledgments: This study was supported by the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) "Nankai Trough Earthquake Research Project for Disaster Mitigation" from fiscal year R2-6.