HypoSVI: Hypocentre inversion with Stein variational inference and physics informed neural networks
- 11 August 2021
- journal article
- research article
- Published by Oxford University Press (OUP) in Geophysical Journal International
- Vol. 228 (1), 698-710
- https://doi.org/10.1093/gji/ggab309
Abstract
We introduce a scheme for probabilistic hypocenter inversion with Stein variational inference. Our approach uses a differentiable forward model in the form of a physics informed neural network, which we train to solve the Eikonal equation. This allows for rapid approximation of the posterior by iteratively optimizing a collection of particles against a kernelized Stein discrepancy. We show that the method is well-equipped to handle highly multimodal posterior distributions, which are common in hypocentral inverse problems. A suite of experiments is performed to examine the influence of the various hyperparameters. Once trained, the method is valid for any seismic network geometry within the study area without the need to build travel time tables. We show that the computational demands scale efficiently with the number of differential times, making it ideal for large-N sensing technologies like Distributed Acoustic Sensing. The techniques outlined in this manuscript have considerable implications beyond just ray-tracing procedures, with the work flow applicable to other fields with computationally expensive inversion procedures such as full waveform inversion.Funding Information
- United States Geological Survey
This publication has 28 references indexed in Scilit:
- Seismic velocity structures in the southern California plate-boundary environment from double-difference tomographyGeophysical Journal International, 2012
- Earthquake Monitoring in Southern California for Seventy-Seven Years (1932-2008)Bulletin of the Seismological Society of America, 2010
- The Fast Marching Method: An Effective Tool for Tomographic Imaging and Tracking Multiple Phases in Complex Layered MediaExploration Geophysics, 2005
- A Reanalysis of the Hypocentral Location and Related Observations for the Great 1906 California EarthquakeBulletin of the Seismological Society of America, 2005
- Finite Difference Traveltime Calculation For Anisotropic MediaGeophysical Journal International, 1993
- Finite difference computation of traveltimes in very contrasted velocity models: a massively parallel approach and its associated toolsGeophysical Journal International, 1991
- The use of a priori data to resolve non-uniqueness in linear inversionGeophysical Journal International, 1979
- Confidence regions and error determinations for seismic event locationReviews of Geophysics, 1965
- The Revision of Earthquake Epicentres, Focal Depths and Origin-Times using a High-Speed ComputerGeophysical Journal International, 1960
- Some Tests of Significance, Treated by the Theory of ProbabilityMathematical Proceedings of the Cambridge Philosophical Society, 1935