Eikonal Tomography With Physics‐Informed Neural Networks: Rayleigh Wave Phase Velocity in the Northeastern Margin of the Tibetan Plateau
Yunpeng Chen, Yunpeng Chen, Sjoerd de Ridder, Sebastian Rost, Zhen Guo, Xiaoyang Wu, Yongshun John Chen, Yongshun John Chen
Abstract
Abstract We present a novel eikonal tomography approach using physics‐informed neural networks (PINNs) for Rayleigh wave phase velocities based on the eikonal equation. The PINN eikonal tomography (pinnET) neural network utilizes deep neural networks as universal function approximators and extracts traveltimes and velocities of the medium during the optimization process. Whereas classical eikonal tomography uses a generic non‐physics based interpolation and regularization step to reconstruct traveltime surfaces, optimizing the network parameters in pinnET means solving a physics constrained traveltime surface reconstruction inversion tackling measurement noise and satisfying physics. We demonstrate this approach by applying it to 25 s surface wave data from ChinArray II sampling the northeastern Tibetan plateau. We validate our results by comparing them to results from conventional eikonal tomography in the same area and find good agreement.