COMPARATIVE ANALYSIS OF DYNAMIC RUPTURE AND STATIC FAULT MODELS FOR TSUNAMI PREDICTION USING T-STOC

抄録

 The modeling of tsunami sources involves various methodologies, particularly in how the earthquake rupture process is represented. Traditional static fault models assume instantaneous tsunami generation, while dynamic rupture models account for the temporal evolution of seismic rupture. This study compares these approaches in simulating tsunami propagation, arrival times, and coastal inundation using numerical simulations of the 2011 Tohoku-Oki tsunami. Six fault models were analyzed, two dynamic rupture models and four static models—using the T-STOC tsunami simulation software. High-resolution bathymetric data from ETOPO1, M7005, and GSI datasets were incorporated to enhance simulation accuracy. Tsunami height time series from NOWPHAS and inundation data from Mori et al. (2011) were used for model validation. Performance evaluation relied on Aida’s Indices and Dynamic Time Warping (DTW) metrics. Results indicate that dynamic rupture models outperform static models in coastal regions, providing better alignment in wave phase and amplitude. While both model types perform comparably in deep water, static models introduce greater nearshore variability and temporal mismatches. However, significant discrepancies in coastal inundation suggest limitations not only in source modeling but also in the inundation models themselves, which may inadequately represent complex terrain and urban features. These findings support the use of dynamic rupture models and highlight the need for improved inundation modeling in tsunami hazard assessments.