Fast GPU-Based Seismogram Simulation From Microseismic Events in Marine Environments Using Heterogeneous Velocity Models

Das, Saptarshi; Chen, Xi; Hobson, M.

doi:10.1109/tci.2017.2654127

Cited by 10 publications

(18 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 1 shows the schematic diagram for the multiple microseismic event detection workflow using Bayesian inference with multimodal nested sampling. Starting from the 3D heterogeneous marine velocity model, we generate the seismograms at the receivers placed at the sea-bed, using an elastic wave propagation solver on Tesla K20 GPUs (Das et al, 2017). Next, the time domain compressed seismic signals are fed to a machine learning algorithm, that is, an ensemble of GP surrogate models for each compressed component using the ARD Matérn 3/2 kernel with linear basis.…”

Section: Bayesian Inferencementioning

confidence: 99%

“…We then perform 4,000 independent simulations of forward wave propagation from random spatial event locations with unit (1 MPa explosive source) amplitude, using an elastic wave equation solver on GPUs, as described in Das et al (2017) and record the seismograms on 23 surface receivers. More details on the source wavelet type, type of measurements and other details of forward simulations are given in Das et al (2017), but here we mainly focus on the datasets for the pressure measurements given by the hydrophones. We then apply a time-domain compression to create a smooth mapping between the event locations and the compressed components of the seismograms.…”

Section: Introductionmentioning

confidence: 99%

“…Making use of this trained fast proxy-based forward model as a means for multi-receiver seismogram generation from random microseismic events, the events' spatial location estimation from given noisy seismograms can be seen as a 3D spatial inference problem using an optimizer or Bayesian sampler with a suitable choice of objective function or likelihood (Tarantola, 2005). However, apart from the spatial locations, the events' amplitudes and origin times are also of interest, and can be simulated with linear scaling and translation operations on unit amplitude seismic responses at random locations, and then combined together for multiple superimposed events as detailed in Das et al (2017). There have been few previous attempts of uncertainty quantification in microseismic location inversion (Eisner et al, 2009;Xuan and Sava, 2010) via Bayesian methods, mostly using P/S-wave arrival time picking (Massin and Malcolm, 2018) for simultaneous estimation of an uncertain velocity model (Zhang et al, 2017), estimation of joint and relative microseismic locations (Poliannikov et al, 2013(Poliannikov et al, , 2014 or even moment tensor estimation (Pugh and White, 2018), etc.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Microseismic event detection in large heterogeneous velocity models using Bayesian multimodal nested sampling

Das

Hobson

Feroz

et al. 2021

DCE

Self Cite

View full text Add to dashboard Cite

In passive seismic and microseismic monitoring, identifying and characterizing events in a strong noisy background is a challenging task. Most of the established methods for geophysical inversion are likely to yield many false event detections. The most advanced of these schemes require thousands of computationally demanding forward elastic-wave propagation simulations. Here we train and use an ensemble of Gaussian process surrogate meta-models, or proxy emulators, to accelerate the generation of accurate template seismograms from random microseismic event locations. In the presence of multiple microseismic events occurring at different spatial locations with arbitrary amplitude and origin time, and in the presence of noise, an inference algorithm needs to navigate an objective function or likelihood landscape of highly complex shape, perhaps with multiple modes and narrow curving degeneracies. This is a challenging computational task even for state-of-the-art Bayesian sampling algorithms. In this paper, we propose a novel method for detecting multiple microseismic events in a strong noise background using Bayesian inference, in particular, the Multimodal Nested Sampling (MultiNest) algorithm. The method not only provides the posterior samples for the 5D spatio-temporal-amplitude inference for the real microseismic events, by inverting the seismic traces in multiple surface receivers, but also computes the Bayesian evidence or the marginal likelihood that permits hypothesis testing for discriminating true vs. false event detection.

show abstract

Section: Bayesian Inferencementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Microseismic event detection in large heterogeneous velocity models using Bayesian multimodal nested sampling

Das

Hobson

Feroz

et al. 2021

DCE

Self Cite

View full text Add to dashboard Cite

show abstract

“…The model specifies values of the propagation velocities for P-and S-waves (V p , V s ) as well as the density ρ of the propagating medium, discretised on a 3D grid of voxels. The solution of the elastic wave equation is a computationally challenging task, which can be accelerated using GPUs (Das et al, 2017). This is implemented in the software K-WAVE (Treeby et al, 2014) Table 1.…”

Section: Simulation Setupmentioning

confidence: 99%

“…Since the solution of the elastic wave equation for forward modelling microseismic traces in complex media is computationally very expensive, this means that, even for small parameter spaces, sampling the posterior distribution becomes extremely challenging or even unattainable. For example, given the geophysical model with microseismic activity considered in Das et al (2017), i.e., a 3D heterogeneous velocity model on a 1 km × 1 km × 3 km grid, the generation of a single seismic trace with a pseudo-spectral method (Treeby et al, 2014) for a given source requires O(1) hour of Graphics Processing Unit (GPU) time with a Nvidia P100 GPU. Using typical MCMC methods, this operation may need to be repeated for tens or hundreds of thousands of points in parameter space, to ensure convergence of the sampling algorithm.…”

mentioning

confidence: 99%

Accelerating Bayesian microseismic event location with deep learning

Mancini¹,

Piras²,

Ferreira³

et al. 2021

Preprint

View full text Add to dashboard Cite

Abstract. We present a series of new open source deep learning algorithms to accelerate Bayesian full waveform point source inversion of microseismic events. Inferring the joint posterior probability distribution of moment tensor components and source location is key for rigorous uncertainty quantification. However, the inference process requires forward modelling of micro-seismic traces for each set of parameters explored by the sampling algorithm, which makes the inference very computationally intensive. In this paper we focus on accelerating this process by training deep learning models to learn the mapping between source location and seismic traces, for a given 3D heterogeneous velocity model, and a fixed isotropic moment tensor for the sources. These trained emulators replace the expensive solution of the elastic wave equation in the inference process. We compare our results with a previous study that used emulators based on Gaussian Processes to invert microseismic events. For fairness of comparison, we train our emulators on the same microseismic traces and using the same geophysical setting. We show that all of our models provide more accurate predictions and ∼100 times faster predictions than the method based on Gaussian Processes, and a O(105) speed-up factor over a pseudo-spectral method for waveform generation. For example, a 2-s long synthetic trace can be generated in ∼10 ms on a common laptop processor, instead of ∼1 hr using a pseudo-spectral method on a high-profile Graphics Processing Units card. We also show that our inference results are in excellent agreement with those obtained from traditional location methods based on travel time estimates. The speed, accuracy and scalability of our open source deep learning models pave the way for extensions of these emulators to generic source mechanisms and application to joint Bayesian inversion of moment tensor components and source location using full waveforms.

show abstract

Towards fast machine-learning-assisted Bayesian posterior inference of microseismic event location and source mechanism

Piras

Mancini

Ferreira

et al. 2022

Geophysical Journal International

View full text Add to dashboard Cite

Summary Bayesian inference applied to microseismic activity monitoring allows the accurate location of microseismic events from recorded seismograms and the estimation of the associated uncertainties. However, the forward modelling of these microseismic events, which is necessary to perform Bayesian source inversion, can be prohibitively expensive in terms of computational resources. A viable solution is to train a surrogate model based on machine learning techniques, to emulate the forward model and thus accelerate Bayesian inference. In this paper, we substantially enhance previous work, which considered only sources with isotropic moment tensors. We train a machine learning algorithm on the power spectrum of the recorded pressure wave and show that the trained emulator allows complete and fast event locations for any source mechanism. Moreover, we show that our approach is computationally inexpensive, as it can be run in less than 1 hour on a commercial laptop, while yielding accurate results using less than 104 training seismograms. We additionally demonstrate how the trained emulators can be used to identify the source mechanism through the estimation of the Bayesian evidence. Finally, we demonstrate that our approach is robust to real noise as measured in field data. This work lays the foundations for efficient, accurate future joint determinations of event location and moment tensor, and associated uncertainties, which are ultimately key for accurately characterising human-induced and natural earthquakes, and for enhanced quantitative seismic hazard assessments.

show abstract

Fast GPU-Based Seismogram Simulation From Microseismic Events in Marine Environments Using Heterogeneous Velocity Models

Cited by 10 publications

References 51 publications

Microseismic event detection in large heterogeneous velocity models using Bayesian multimodal nested sampling

Microseismic event detection in large heterogeneous velocity models using Bayesian multimodal nested sampling

Accelerating Bayesian microseismic event location with deep learning

Towards fast machine-learning-assisted Bayesian posterior inference of microseismic event location and source mechanism

Contact Info

Product

Resources

About