A Quantitative Comparison and Validation of Finite‐Fault Models: The 2011 Tohoku‐Oki Earthquake

• Published in: Journal of geophysical research. Solid earth Vol. 129; no. 10
• Main Authors: Wong, Jeremy Wing Ching, Fan, Wenyuan, Gabriel, Alice‐Agnes
• Format: Journal Article
• Language: English
• Published: 01.10.2024
• Subjects: finite‐fault models; source kinematics; Tohoku‐Oki earthquake
• ISSN: 2169-9313; 2169-9356
• DOI: 10.1029/2024JB029212

Large earthquakes rupture faults over hundreds of kilometers within minutes. Finite‐fault models image these processes and provide observational constraints for understanding earthquake physics. However, finite‐fault inversions are subject to non‐uniqueness and uncertainties. The diverse range of published models for the well‐recorded 2011 Mw ${M}_{w}$ 9.0 Tohoku‐Oki earthquake illustrates this challenge, and its rupture process remains under debate. Here, we comprehensively compare 32 published finite‐fault models of the Tohoku‐Oki earthquake. We aim to identify the most coherent slip features of the Tohoku‐Oki earthquake from these slip models and develop a new method for quantitatively analyzing their variations. We find that the models correlate poorly at 1‐km subfault size, irrespective of the data type. In contrast, model agreement improves significantly with increasing subfault sizes, consistently showing that the largest slip occurs up‐dip of the hypocenter near the trench. We use the set of models to test the sensitivity of available teleseismic, regional seismic, and geodetic observations. For the large Tohoku‐Oki earthquake, we find that the analyzed finite‐fault models are less sensitive to slip features smaller than 64 km. When we use the models to compute synthetic seafloor deformation, we observe strong variations in the synthetics, suggesting their sensitivity to small‐scale slip features. Our newly developed approach offers a quantitative framework to identify common features in distinct finite‐fault slip models and to analyze their robustness using regional and global geophysical observations for megathrust earthquakes. Our results indicate that dense offshore instrumentation is critical for resolving the rupture complexities of megathrust earthquakes. Plain Language Summary Large earthquakes often rupture in unexpected ways across extensive areas of faults. Scientists use finite‐fault models to resolve these processes in detail. These models use different observations to help us understand earthquakes and plan for future hazard mitigation and risk management. However, these models are not perfect: they are often challenging to resolve, and different models of the same earthquake can show very different results. For example, many different models have been published for the 2011 Mw ${M}_{w}$ 9.0 Tohoku‐Oki earthquake, each showing varying “slip features” of how the megathrust moved during the same event. In this study, we quantitatively compare 32 of these models of this earthquake with each other and with observations in a new and systematic way. The models show coherent features at a scale of 64 km while disagreeing on the smaller, fine‐scale details. We find that such fine‐scale features of the earthquake cannot be uniquely resolved by the commonly used remote observations, such as geodetic, regional seismic‐geodetic, teleseismic, and tsunami data. Our study suggests that to better understand large megathrust earthquakes, dense networks of instruments placed directly offshore close to the megathrust are needed for robustly resolving their rupture processes. Key Points We evaluate 32 finite‐fault models of the 2011 Tohoku‐Oki earthquake, using realistic slab geometry and varying spatial resolution The models show strong agreement on large‐scale slip features, but significant variability arises from small‐scale secondary slip features We evaluate the ability to resolve model variability using geodetic, regional seismic, teleseismic, and offshore uplift observational data