Abstract
Summary Seismic phase detection, identification and first-onset picking are basic but essential routines to analyse earthquake data. As both the number of seismic stations, globally and regionally, and the number of experiments greatly increase due to ever greater availability of instrumentation, automated data processing becomes more and more essential. E.g., for modern seismic experiments involving 100s to even 1,000s instruments, conventional human analyst-based identification and picking of seismic phases is becoming unfeasible, and the introduction of automatic algorithms mandatory. In this paper, we introduce DeepPhasePick, an automatic two-stage method that detects and picks P and S seismic phases from local earthquakes. The method is entirely based on highly optimized deep neural networks, consisting of a first stage that detects the phases using a convolutional neural network, and a second stage that uses two recurrent neural networks to pick both phases. Detection is performed on three-component seismograms. P- and S-picking is then conducted on the vertical and the two-horizontal components, respectively. Systematic hyperparameter optimization was applied to select the best model architectures and to define both the filter applied to preprocess the seismic data as well as the characteristics of the window sample used to feed the models. We trained DeepPhasePick using seismic records extracted from two sets of manually-picked event waveforms originating from northern Chile (∼39,000 records for detection and ∼36,000 records for picking). In different tectonic regimes, DeepPhasePick demonstrated the ability to both detect P and S phases from local earthquakes with high accuracy, as well as predict P- and S-phase time onsets with an analyst level of precision. DeepPhasePick additionally computes onset uncertainties based on the Monte Carlo Dropout technique as an approximation of Bayesian inference. This information can then further feed an associator algorithm in an earthquake location procedure.
Highlights
One of the most fundamental components in any earthquake hypocentre estimation routine is the identification and picking of the arriving P and S seismic phases
The accurate predictions produced by DeepPhasePick result from the highly optimized set of hyperparameters defining its convolutional and recurrent deep neural networks trained for the tasks of seismic phase detection and picking, respectively
Predictions performed on seismic samples from two independent test sets show that DeepPhasePick is capable of recognizing manually as well as automatically picked P and S phases with high accuracy, it decreases for lower-quality automatic picks. These results demonstrate that DeepPhasePick predicts phase time onsets which are comparable to those picked by analysts, as can be seen from the narrow time residual distributions in Figs 11(c) and 12(c)
Summary
One of the most fundamental components in any earthquake hypocentre estimation routine is the identification and picking of the arriving P and S seismic phases. The phase picking task was commonly performed manually by analysts, who identified each phase arrival based on their training and experience. As the available seismic data has rapidly increased over time, the use of automatic phase detection algorithms has become increasingly necessary. These automatic algorithms encompass detectors which are based on the energy or frequency content of the seismic waveforms such as STA/LTA Phase detectors based on frequency or energy have been used in the past as part of multistage automatic earthquake location procedures that allowed the creation of high-quality earthquake These automatic algorithms encompass detectors which are based on the energy or frequency content of the seismic waveforms such as STA/LTA (e.g. Allen 1978; Baer & Kradolfer 1987; Earle & Shearer 1994; Sleeman & van Eck 1999; Aldersons 2004; Di Stefano et al 2006; Diehl et al 2009), those based on correlations of template waveforms against continuous seismic data (e.g. Van Trees 1968; Harris 1991; Gibbons & Ringdal 2006), and detectors based on the representation of seismic data as a linear combination of orthogonal basis waveforms (Scharf & Friedlander 1994; Harris 1997, 2001).
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
