Discrimination of Seismic Signals from Earthquakes and Tectonic Tremor by Applying a Convolutional Neural Network to Running Spectral Images

Sugiyama, Daisuke; Hori, Takane; Kuwatani, Tatsu; Tsuboi, Seiji

doi:10.1785/0220180279

Cited by 51 publications

(44 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Very recently an improved Convolutional Neural Network method was applied on running spectral images to identify tectonic tremor, local earthquake, and noise by Nakano et al (). They successfully obtained a high accuracy up to 99.5% using all components and stations available in the tremor zone.…”

Section: Discussionmentioning

confidence: 99%

“…This challenge was met here as the Tremor in Taiwan being highly similar with Noise in amplitude (Chuang et al, ) and spectral behavior (Figure ). Different from Nakano et al () where the features selection was not conducted, this study attempts to select the best features using the Fisher's criterion.…”

Section: Discussionmentioning

confidence: 99%

“…The previous methods of single station detection in Cascadia focus only on the amplitude for tremor passband (1–5 Hz; Brudzinski & Allen, ; Kao et al, ; Sit et al, ), where the detection criteria heavily depends on station quality and signal‐to‐noise ratio. Using Convolutional Neural Network, Nakano et al () successfully discriminated tremors from local earthquakes and noise in spectral images. Which seismic features exactly are needed for tremor recognition remain unexplored?…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Investigation of Single‐Station Classification for Short Tectonic Tremor in Taiwan

et al. 2019

View full text Add to dashboard Cite

Tectonic tremor in Taiwan is known for its short duration and weak amplitude. Recent observations from the mountain area (Central Range) allows us to document such type of tectonic tremor recorded at a close distance and to investigate a single station technique for a minute‐long tremor. To effectively distinguish tremor from earthquakes and noise, we applied the Fisher's class separability criterion to automatically select the optimal feature subset from a set of commonly adopted feature candidates. During the study period of 1 January to 16 September 2016 when a local seismic array was deployed, we successfully differentiated tremor from local earthquakes and noise with high accuracy of 86.6% to 98.9% from three stations using a k‐nearest neighbors classifier. Other than the maximum amplitude, number of peaks, and energy of the 2‐ to 8‐Hz passband, the spikiness of discrete Fourier transforms median in time is also found to be important to separate tremor from noise.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Investigation of Single‐Station Classification for Short Tectonic Tremor in Taiwan

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Additionally, the performance of our similarity-based discriminator is directly compared to that of two state-of-the-art methods: the SVM-based discriminator proposed in (Kortstrm et al, 2016) and the SRSpec-CNN discriminator adapted from the work of (Nakano et al, 2019). In particular, our SVM and CNN implementations both utilize the full 149,036 training waveforms from the training set.…”

Section: Evaluation Criteriamentioning

confidence: 99%

Beyond Correlation: A Path‐Invariant Measure for Seismogram Similarity

Dickey

Borghetti

Junek

et al. 2019

Seismological Research Letters

View full text Add to dashboard Cite

Similarity search is a popular technique for seismic signal processing, with template matching, matched filters and subspace detectors being utilized for a wide variety of tasks, including both signal detection and source discrimination. Traditionally, these techniques rely on the cross-correlation function as the basis for measuring similarity. Unfortunately, seismogram correlation is dominated by path effects, essentially requiring a distinct waveform template along each path of interest. To address this limitation, we propose a novel measure of seismogram similarity that is explicitly invariant to path. Using Earthscope's USArray experiment, a path-rich dataset of 207,291 regional seismograms across 8,452 unique events is constructed, and then employed via the batch-hard triplet loss function, to train a deep convolutional neural network which maps raw seismograms to a low dimensional embedding space, where nearness on the space corresponds to nearness of source function, regardless of path or recording instrumentation. This path-agnostic embedding space forms a new representation for seismograms, characterized by robust, source-specific features, which we show to be useful for performing both pairwise event association as well as templatebased source discrimination with a single template.

show abstract

“…This innovative technology has had a significant impact not only on research but also on people's daily lives such as language translation and automatic driving. Applications of deep learning to seismology are also proceeding rapidly, including the detection of P-and S-wave arrival times (Zhu and Beroza 2018), determination of P-wave arrival times and first-motion polarities (Ross et al 2018), detection and location determination of earthquakes (Perol et al 2018), prediction of aftershock distributions (DeVries et al 2018), and discrimination of seismic signals from earthquakes and tectonic tremors (Nakano et al 2019).…”

Section: Introductionmentioning

confidence: 99%

P-wave first-motion polarity determination of waveform data in western Japan using deep learning

Hara

Fukahata

Iio

2019

Earth Planets Space

View full text Add to dashboard Cite

P-wave first-motion polarity is the most useful information in determining the focal mechanisms of earthquakes, particularly for smaller earthquakes. Algorithms have been developed to automatically determine P-wave first-motion polarity, but the performance level of the conventional algorithms remains lower than that of human experts. In this study, we develop a model of the convolutional neural networks (CNNs) to determine the P-wave first-motion polarity of observed seismic waveforms under the condition that P-wave arrival times determined by human experts are known in advance. In training and testing the CNN model, we use about 130 thousand 250 Hz and about 40 thousand 100 Hz waveform data observed in the San-in and the northern Kinki regions, western Japan, where three to four times larger number of waveform data were obtained in the former region than in the latter. First, we train the CNN models using 250 Hz and 100 Hz waveform data, respectively, from both regions. The accuracies of the CNN models are 97.9% for the 250 Hz data and 95.4% for the 100 Hz data. Next, to examine the regional dependence, we divide the waveform data sets according to the observation region, and then we train new CNN models with the data from one region and test them using the data from the other region. We find that the accuracy is generally high (95%) and the regional dependence is within about 2%. This suggests that there is almost no need to retrain the CNN model by regions. We also find that the accuracy is significantly lower when the number of training data is less than 10 thousand, and that the performance of the CNN models is a few percentage points higher when using 250 Hz data compared to 100 Hz data. Distribution maps, on which polarities determined by human experts and the CNN models are plotted, suggest that the performance of the CNN models is better than that of human experts.

show abstract

Discrimination of Seismic Signals from Earthquakes and Tectonic Tremor by Applying a Convolutional Neural Network to Running Spectral Images

Cited by 51 publications

References 37 publications

Investigation of Single‐Station Classification for Short Tectonic Tremor in Taiwan

Investigation of Single‐Station Classification for Short Tectonic Tremor in Taiwan

Beyond Correlation: A Path‐Invariant Measure for Seismogram Similarity

P-wave first-motion polarity determination of waveform data in western Japan using deep learning

Contact Info

Product

Resources

About