arXiv subjects

Ryoichiro Agata

Publications and source records attributed to Ryoichiro Agata.

7 recordsLinked to original sources

arXiv2026

Bayesian three-dimensional seismic travel-time tomography for active- and passive-source seismic data using physics-informed neural network

Accurate 3D seismic velocity modeling through seismic travel-time tomography using both active- and passive-source data provides critical underpinning models for seismicity monitoring and hazard assessment. Because travel-time tomography is an inherently ill-posed inverse problem, UQ of the estimated models using Bayesian methods is also important for reliable downstream interpretations and analyses. However, Bayesian inference for 3D tomography based on conventional grid-based representations faces the ``curse of dimensionality'' and severe computational bottlenecks. Consequently, rigorous Bayesian UQ for margin-wide 3D travel-time tomography has remained largely unexplored. In this study, we propose a meshless 3D Bayesian travel-time tomography method that combines PINNs with a neural representation of the velocity structure, enabling tractable and data-efficient Bayesian inference through function-space particle-based variational inference. To efficiently integrate passive-source data into the Bayesian estimation of the velocity structure, we conduct analytical marginalization treating uncertain source parameters as nuisance parameters, with passive-source relocation carried out in post-processing. We validated the capability of our approach for 3D problems through synthetic experiments. Furthermore, we applied the method to a real-world dataset from marine active-source surveys and natural earthquakes off the Kii Peninsula, Nankai Trough. Our probabilistic 3D ensemble successfully resolves key geological features and provides data-consistent uncertainty maps. The posterior mean hypocenters shifted mainly in the vertical direction by 10-15 km, consistent with a previous relocation result. Finally, the neural representation drastically reduces storage requirements for the entire ensemble velocity model, highlighting the scalability and data efficiency of the proposed framework.

physics.geo-ph↗

arXiv2026

Functional-prior-based approaches to Bayesian PDE-constrained inversion using physics-informed neural networks

Physics-informed neural networks (PINNs) provide a mesh-free framework for solving PDE-constrained inverse problems, but their extension to Bayesian inversion still faces a fundamental difficulty: prior distributions are typically defined in the weight space of neural networks, whereas physically meaningful prior assumptions are more naturally expressed in function space. In this study, we introduce a unified framework, termed functional-prior-based approaches to Bayesian PDE-constrained inversion using physics-informed neural networks (fpBPINN), to incorporate functional priors into Bayesian PINN-based inversion. We consider two complementary approaches. The first is a functional-prior-informed Bayesian PINN (FPI-BPINN), in which a neural network weight prior is learned to be consistent with a prescribed functional prior, and Bayesian inference is subsequently performed in weight space. The second is function-space particle-based variational inference for PINNs (fParVI-PINN), which performs Bayesian estimation using ParVI directly in function space. We also show that random Fourier features (RFF) play an important role in representing Gaussian functional priors with neural networks and in improving posterior approximation. We applied the proposed approaches to one-dimensional seismic traveltime tomography and two-dimensional Darcy-flow permeability inversion. These numerical experiments showed that both approaches accurately estimated posterior distributions, highlighting the significance of introducing physically interpretable functional priors into Bayesian PINN-based inverse problems. We also identified the contrasting advantages of FPI-BPINN and fParVI-PINN, namely flexibility and accuracy, respectively.

physics.geo-ph↗

arXiv2025

Three-dimensional crustal deformation analysis using physics-informed deep learning

Earthquake-related phenomena such as seismic waves and crustal deformation impact broad regions, requiring large-scale modeling with careful treatment of artificial outer boundaries. Physics-informed neural networks (PINNs) have been applied to analyze wavefront propagation, acoustic and elastic waveform propagations, and crustal deformation in semi-infinite domains. In this study, we investigated the capability of PINNs for modeling earthquake crustal deformation in 3-D structures. To improve modeling accuracy, four neural networks were constructed to represent the displacement and stress fields in two subdomains divided by a fault surface and its extension. Forward simulations exhibited high accuracy for internal deformation but yielded errors for rigid motions, underscoring the inherent difficulty in constraining static deformation at an infinite distance. In the inversion analysis, fault slip distributions were estimated using surface observational data. Application to real data from the 2008 Iwate-Miyagi inland earthquake showed a fault slip consistent with previous studies, despite underestimation of the magnitude. This study demonstrates the capability of PINNs to analyze 3-D crustal deformation, thereby offering a flexible approach for large-scale earthquake modeling using real-world observations and crustal structures.

physics.geo-ph↗

arXiv2024

Quantification of Uncertainty and Its Propagation in Seismic Velocity Structure and Earthquake Source Inversion

In earthquake source inversions aimed at understanding diverse fault activities on earthquake faults using seismic observation data, uncertainties in velocity structure models are typically not considered. As a result, biases and underestimations of uncertainty can occur in source inversion. This article provides an overview of the author's efforts to address this issue by quantitatively evaluating the uncertainty in velocity structure models and appropriately accounting for its propagation into source inversion. First, the Bayesian multi-model source inversion method that can incorporate such uncertainties as probability distributions in the form of ensembles is explained. Next, a Bayesian traveltime tomography technique utilizing physics-informed neural networks (PINN) to quantify uncertainties in velocity structure models is introduced. Furthermore, the author's recent efforts to integrate these methods and apply them to hypocenter determination in the Nankai Trough region are briefly discussed. The article also outlines future prospects of source inversions considering uncertainties in velocity structure models and the anticipated role of the emerging scientific machine learning (SciML) methods such as PINN.

physics.geo-ph↗

arXiv2024

HypoNet Nankai: Rapid hypocenter determination tool for the Nankai Trough subduction zone using physics-informed neural networks

Accurate hypocenter determination in the Nankai Trough subduction zone is essential for hazard assessment and advancing our understanding of seismic activity in the region. A handy hypocenter determination tool incorporating a realistic 3D velocity structure, accessible to the scientific community, is beneficial. In this study, we developed HypoNet Nankai, a rapid hypocenter determination tool based on a physics-informed neural network (PINN) emulator (surrogate model) for travel time calculations. This tool leverages a PINN trained to predict P-wave travel times between arbitrary underground sources and surface receivers with a realistic 3D P-wave velocity structure model of the Nankai Trough subduction zone that incorporates marine seismic survey data. The PINN embeds physical laws, namely, the Eikonal equation, directly into the loss function of training and circumvents the need for labeled training data. To address the training challenges posed by small-scale features in the velocity model, we employed a simple domain decomposition approach and Fourier feature embedding. Once trained, the PINN immediately infers the P-wave travel time, enabling rapid hypocenter determination. The data size required to store NNs for travel time calculations is significantly smaller than those of conventional travel-time tables. HypoNet Nankai provides high flexibility for addition of new observation points. We verified HypoNet Nankai by comparing its performance with a widely used grid-based numerical method for forward travel time calculations and synthetic hypocenter determination. In both tests, HypoNet Nankai provided results consistent with those for the conventional method. HypoNet Nankai offers a rapid, accurate, and easy-to-use hypocenter determination method for the Nankai Trough subduction zone, with greater data efficiency and extendibility compared to conventional approaches.

physics.geo-ph↗

arXiv2024

Physics-informed deep learning quantifies propagated uncertainty in seismic structure and hypocenter determination

Subsurface seismic velocity structure is essential for earthquake source studies, including hypocenter determination. Conventional hypocenter determination methods ignore the inherent uncertainty in seismic velocity structure models, and the impact of this oversight has not been thoroughly investigated. Here, we address this issue by employing a physics-informed deep learning (PIDL) approach that quantifies uncertainty in seismic velocity structure modeling and its propagation to hypocenter determination by introducing neural network ensembles trained on active seismic survey data, earthquake observation data, and the physical equation of wavefront movement. An analysis of an earthquake in southwest Japan using our method revealed that accounting for such uncertainty propagation significantly reduced the bias and uncertainty underestimation in the hypocenter determination, enabling quantitative evaluation of the focal depth relative to the plate boundary. Our results highlight the potential of PIDL for various geophysical inverse problems, such as investigating earthquake source parameters, which inherently suffer from uncertainty propagation.

physics.geo-ph↗

arXiv2023

Bayesian seismic tomography based on velocity-space Stein variational gradient descent for physics-informed neural network

In this study, we propose a Bayesian seismic tomography inference method using physics-informed neural networks (PINN). PINN represents a recent advance in deep learning, offering the possibility to enhance physics-based simulations and inverse analyses. PINN-based deterministic seismic tomography uses two separate neural networks (NNs) to predict seismic velocity and travel time. Naive Bayesian NN (BNN) approaches are unable to handle the high-dimensional spaces spanned by the weight parameters of these two NNs. Hence, we reformulate the problem to perform the Bayesian estimation exclusively on the NN predicting seismic velocity, while the NN predicting travel time is used only for deterministic travel time calculations, with the help of the adjoint method. Furthermore, we perform BNN by introducing a function-space Stein variational gradient descent (SVGD), which performs particle-based variational inference in the space of the function predicted by the NN (i.e., seismic velocity), instead of in the traditional weight space. The result is a velocity-space SVGD for the PINN-based seismic tomography model (vSVGD-PINN-ST) that decreases the complexity of the problem thus enabling a more accurate and physically consistent Bayesian estimation, as confirmed by synthetic tests in one- and two-dimensional tomographic problem settings. The method allows PINN to be applied to Bayesian seismic tomography practically for the first time. Not only that, it can be a powerful tool not only for geophysical but also for general PINN-based Bayesian estimation problems associated with compatible NNs formulations and similar, or reduced, complexity.

physics.geo-ph↗