[S21] AM-1

[S21-01] Convolutional Neural Networks for Seismic Velocity Model Building and Uncertainty Quantification

*Fan Yu¹, Ehsan Jamali Hondori², Jin-Oh Park¹ (1. Atmosphere and Ocean Research Institute, The University of Tokyo, 2. Geoscience Enterprise Inc.)

Velocity is a fundamental parameter in seismology, essential for studying the physical properties of subsurface and converting time-recorded seismograms into representative depth sections in multi-channel seismic reflection studies. Practically, the primary output of velocity model building is a base subsurface model utilized for seismic imaging and interpretation workflows. Conventional processes to develop a seismic velocity model (e.g., traveltime tomography or full waveform inversion) are generally operator-dependent, time-consuming, and often require a reasonably accurate initial velocity model to converge to a stable solution. To address these challenges, we propose a method to estimate the interval velocity models faster and more accurately by using Convolutional Neural Networks (CNNs). In the recently developed velocity model building tools, CNNs have proven to be effective for simple geological settings (Araya-Polo et al., 2018; Simon et al., 2023). However, in more complex geologies such as subduction systems, CNNs have a gap compared to traditional model building methods. In addition to training the CNNs model with the simple layered geological settings, we implemented a more complex synthetic geological settings into the training process to solve this issue. Our CNNs model uses the residual move-out semblance panel as input data for training and the labels for each input are from the corresponding interval velocity profile at the same location. In this study, Monte-Carlo Dropout is implemented to perform uncertainty quantification of the predicted P-wave velocity model. Our CNNs model is trained by synthetic data and then tested on real seismic data after the transfer learning.