GEOSCAN Search Results: Fastlink


TitleCombining deep learning and the source-scanning algorithm for improved seismic monitoring
AuthorDokht, R M HORCID logo; Kao, HORCID logo; Ghofrani, HORCID logo; Visser, R
SourceBulletin of the Seismological Society of America 2022 p. 1-15,
Alt SeriesNatural Resources Canada, Contribution Series 20220075
PublisherSeismological Society of America
Mediapaper; digital; on-line
File formatpdf; html
SubjectsScience and Technology; Nature and Environment; seismology; seismic data; earthquakes
Illustrationsgraphs; diagrams; charts
ProgramEnvironmental Geoscience Shale Gas - induced seismicity
Released2022 06 23
In the present study, we develop an integrated framework for simultaneous detection of seismic events and picking phase arrival times, phase association, and locating earthquakes. The proposed model combines the accuracy of convolutional neural networks for classification tasks and the efficiency of waveform-based algorithms for identifying coherent seismic arrivals. We find that our model strongly dominates the classic techniques, especially in identifying small magnitude earthquakes. We apply our model to one month of continuous seismic data recorded in western Canada for monitoring seismic activity associated with fluid injection operations. In comparison with previously developed deep-learning models, our technique reveals a nearly identical performance without human interaction during the entire process of picking the phase arrival times and locating the associated events.
Summary(Plain Language Summary, not published)
Due to the successful application of machine learning techniques in the previous seismic studies, we designed and trained an Artificial Neural Network to identify the major seismic arrivals from earthquakes in the time-frequency domain. The proposed model overcomes the challenges of traditional earthquake detection techniques, especially in identifying small-magnitude earthquakes, and enables us to obtain reliable earthquake locations without user interaction. The parallel implementation of the proposed framework allows us to tackle the large dataset problems, which is necessary for real-time seismic monitoring.

Date modified: