Semi-supervised Surface Wave Tomography with Wasserstein Cycle-consistent GAN: Method and Application on Southern California Plate Boundary Region

Abstract

A machine learning based method is developed for 1-D shear wave velocity (Vs) inversion to include observed dispersion data into the training process • The Wasserstein Cycle-GAN algorithm is used to improve training stability and spatial continuity of the output 3-D Vs model • The final Vs model shows reasonable data misfits, sharper images of major faults, and is consistent with the largescale surface geology

Autocorrelation R2 on Mars

“…Seismic interferometry has been widely applied to retrieve the empirical Green's functions of surface waves (Berg et al., 2018; Miao et al., 2022; Qiu et al., 2020, 2021; Schimmel et al., 2018; Shen et al., 2013; Yao et al., 2006) and body waves (Clayton, 2020; Feng et al., 2017, 2021; Gorbatov et al., 2013; Kennett, 2015; Oren & Nowack, 2017; She et al., 2022) from ambient noise auto‐ and cross‐correlations. Many ambient noise surface wave tomography studies on Earth have focused on the calculations of the Rayleigh wave phase velocity to invert for the S‐wave velocity of sedimentary basins (Cai et al., 2022; Hannemann et al., 2014; Pan et al., 2016; Qiu et al., 2019; Shirzad & Shomali, 2014), the crust and upper mantle (Li et al., 2012; Lin et al., 2014; Nguyen et al., 2022; Yao et al., 2008; Zhang et al., 2018). Global tomography analysis suggested that the long period Rayleigh waves (e.g., >100 s) can be recovered by cross‐correlations of the Earth's hum (Haned et al., 2016; Nishida et al., 2009).…”

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

“…Most studies in this category design deep neural networks that are capable to capture the complex transformation from the measured data space to the desired model parameter space, train these machines using paired models and their corresponding synthetic data, and apply the trained machines to field datasets. Applications of such a framework range across the whole spectrum of geophysical inverse problems, including surface wave dispersion inversion and tomography (Cai et al., 2022; X. Zhang & Curtis, 2021), seismic‐to‐petrophysics inversion (Xiong et al., 2021; C. Zou et al., 2021), crustal thickness and Vp / Vs estimation from receiver functions (F. Wang et al., 2022), earthquake and microseismicity moment tensor inversion (Chen et al., 2022; Steinberg et al., 2021), magnetic, gravity, and ground‐penetrating radar (GPR) data inversion (R. Huang et al., 2021; Leong & Zhu, 2021; Nurindrawati & Sun, 2020), and thermal evolution estimation for Mars (Agarwal et al., 2021). Y. Wu et al.…”

Characterization of Subsurface Hydrogeological Structures With Convolutional Conditional Neural Processes on Limited Training Data

“…With the extraordinary ability to extract complex characteristics of deep learning technique, various generative models are applied to hydrogeological modeling (Cai et al., 2022; Laloy et al., 2018; Lopez‐Alvis et al., 2022; Yang et al., 2022; Zhan et al., 2022; K. Zhang et al., 2021). In the majority of relevant research, generative models are trained to learn the distribution of training data and reproduce the probability distribution according to the test data.…”