Title | Seismic event and phase detection using time-frequency representation and convolutional neural networks |
| |
Author | Dokht, R M H ;
Kao, H ; Visser, R; Smith, B |
Source | Seismological Research Letters 2019 p. 1-10, https://doi.org/10.1785/0220180308 |
Image |  |
Year | 2019 |
Alt Series | Natural Resources Canada, Contribution Series 20180263 |
Publisher | Seismological Society of America (SSA) |
Document | serial |
Lang. | English |
Media | paper; on-line; digital |
File format | pdf; html |
Subjects | geophysics; fossil fuels; seismology; earthquakes; array seismology; seismicity; seismic waves; p waves; s waves; models; earthquake catalogues; earthquake magnitudes; seismic risk; earthquake risk;
petroleum industry; hydrocarbon recovery; Methodology |
Illustrations | location maps; histograms; graphs; tables; frequency distribution diagrams |
Program | Environmental Geoscience Shale Gas - induced seismicity |
Released | 2019 01 16 |
Abstract | The availability of abundant digital seismic records and successful application of deep learning in pattern recognition and classification problems enable us to achieve a reliable earthquake detection
framework. To overcome the limitations and challenges of conventional methods, which are mainly due to an incomplete set of template waveforms and low signal-to-noise ratio, we design a generalized model to improve discrimination between earthquake
and noise recordings using a deep convolutional network (ConvNet). Exclusively based on a dataset of over 4900 earthquakes recorded over a period of 3 yrs in western Canada, a multilayer ConvNet is trained to learn general characteristics of
background noise and earthquake signals in the time-frequency domain. In the next step, we train a secondary network using the wavelet transform of the major seismic arrivals to separate P from S waves and estimate their approximate arrival times.
The results of validation experiments demonstrate promising performance and achieve an average accuracy of nearly 99% for both networks. To investigate the applicability of our algorithm, we apply the trained model on an independent dataset recently
recorded in northeastern British Columbia (NE BC). It is found that deep-learning-based methods are superior to traditional techniques in detecting a higher number of seismic events at significantly less computational cost. |
Summary | (Plain Language Summary, not published) We design a generalized model to improve the discrimination between earthquake signals and background noise using an artificial intelligence (AI) based
approach. We train this model with seismograms recorded for over 4900 events over a period of 3 years in western Canada. A secondary module is designed to separate P from S waves and to estimate their arrival times. The results of validation
experiments demonstrate promising performance and achieve an average accuracy of nearly 99%. It is concluded that deep learning-based methods are superior to traditional techniques in detecting a higher number of seismic events at significantly less
computational cost. |
GEOSCAN ID | 313022 |
|
|