date: 2026-04-10T12:28:57Z
pdf:PDFVersion: 1.4
pdf:docinfo:title: DeepSubDAS: an earthquake phase picker from submarine distributed acoustic sensing data
xmp:CreatorTool: OUP
access_permission:can_print_degraded: true
subject: DOI: 10.1093/gji/ggag061 Geophysical Journal International, 245, 2, 9-2-2026. Abstract: Given the scarcity of seismometers in marine environments, traditional seismology has limited effectiveness in oceanic regions. Submarine Distributed Acoustic Sensing (DAS) systems offer a promising alternative for seismic monitoring in these areas. However, the existing machine learning model trained on land-based DAS data does not perform well with submarine DAS due to differences in noise characteristics, deployment conditions and environmental factors. This study presents a machine learning approach tailored specifically to submarine DAS data to enable automated seismic event detection and P- and S-wave identification. Leveraging DeepLab v3, a neural network architecture optimized for semantic segmentation, we developed a specialized model to handle the unique challenges of submarine DAS data. Our model was trained and validated on a data set comprising nearly 57 million manually and semi-automatically labelled seismic records from multiple globally distributed submarine sites, providing a robust basis for accurate seismic detection. The model adapts to a variety of deployment scenarios and can process DAS data from cables with different lengths, configurations and channel spacings, making it versatile for various ocean environments. We thus provide an adaptable and efficient tool for automated earthquake analysis of DAS data, which has the potential to enhance real-time earthquake monitoring and tsunami early warning in submarine environments.
language: English
dc:format: application/pdf; version=1.4
pdf:docinfo:creator_tool: OUP
access_permission:fill_in_form: true
pdf:encrypted: false
dc:title: DeepSubDAS: an earthquake phase picker from submarine distributed acoustic sensing data
modified: 2026-04-10T12:28:57Z
cp:subject: DOI: 10.1093/gji/ggag061 Geophysical Journal International, 245, 2, 9-2-2026. Abstract: Given the scarcity of seismometers in marine environments, traditional seismology has limited effectiveness in oceanic regions. Submarine Distributed Acoustic Sensing (DAS) systems offer a promising alternative for seismic monitoring in these areas. However, the existing machine learning model trained on land-based DAS data does not perform well with submarine DAS due to differences in noise characteristics, deployment conditions and environmental factors. This study presents a machine learning approach tailored specifically to submarine DAS data to enable automated seismic event detection and P- and S-wave identification. Leveraging DeepLab v3, a neural network architecture optimized for semantic segmentation, we developed a specialized model to handle the unique challenges of submarine DAS data. Our model was trained and validated on a data set comprising nearly 57 million manually and semi-automatically labelled seismic records from multiple globally distributed submarine sites, providing a robust basis for accurate seismic detection. The model adapts to a variety of deployment scenarios and can process DAS data from cables with different lengths, configurations and channel spacings, making it versatile for various ocean environments. We thus provide an adaptable and efficient tool for automated earthquake analysis of DAS data, which has the potential to enhance real-time earthquake monitoring and tsunami early warning in submarine environments.
pdf:docinfo:subject: DOI: 10.1093/gji/ggag061 Geophysical Journal International, 245, 2, 9-2-2026. Abstract: Given the scarcity of seismometers in marine environments, traditional seismology has limited effectiveness in oceanic regions. Submarine Distributed Acoustic Sensing (DAS) systems offer a promising alternative for seismic monitoring in these areas. However, the existing machine learning model trained on land-based DAS data does not perform well with submarine DAS due to differences in noise characteristics, deployment conditions and environmental factors. This study presents a machine learning approach tailored specifically to submarine DAS data to enable automated seismic event detection and P- and S-wave identification. Leveraging DeepLab v3, a neural network architecture optimized for semantic segmentation, we developed a specialized model to handle the unique challenges of submarine DAS data. Our model was trained and validated on a data set comprising nearly 57 million manually and semi-automatically labelled seismic records from multiple globally distributed submarine sites, providing a robust basis for accurate seismic detection. The model adapts to a variety of deployment scenarios and can process DAS data from cables with different lengths, configurations and channel spacings, making it versatile for various ocean environments. We thus provide an adaptable and efficient tool for automated earthquake analysis of DAS data, which has the potential to enhance real-time earthquake monitoring and tsunami early warning in submarine environments.
pdf:docinfo:creator: Xiao Han, Tilmann Frederik, van den Ende Martijn, Rivet Diane, Loureiro Afonso, Tsuji Takeshi, Ugalde Arantza, Shi Qibin, Denolle Marine A.
meta:author: Xiao Han, Tilmann Frederik, van den Ende Martijn, Rivet Diane, Loureiro Afonso, Tsuji Takeshi, Ugalde Arantza, Shi Qibin, Denolle Marine A.
meta:creation-date: 2026-03-10T09:43:56Z
created: 2026-03-10T09:43:56Z
access_permission:extract_for_accessibility: true
Creation-Date: 2026-03-10T09:43:56Z
pdf:docinfo:custom:doi: 10.1093/gji/ggag061
Author: Xiao Han, Tilmann Frederik, van den Ende Martijn, Rivet Diane, Loureiro Afonso, Tsuji Takeshi, Ugalde Arantza, Shi Qibin, Denolle Marine A.
producer: Acrobat Distiller 25.0 (Windows); modified using iTextSharp.LGPLv2.Core 3.7.4.0
pdf:docinfo:producer: Acrobat Distiller 25.0 (Windows); modified using iTextSharp.LGPLv2.Core 3.7.4.0
doi: 10.1093/gji/ggag061
pdf:unmappedUnicodeCharsPerPage: 1
dc:description: DOI: 10.1093/gji/ggag061 Geophysical Journal International, 245, 2, 9-2-2026. Abstract: Given the scarcity of seismometers in marine environments, traditional seismology has limited effectiveness in oceanic regions. Submarine Distributed Acoustic Sensing (DAS) systems offer a promising alternative for seismic monitoring in these areas. However, the existing machine learning model trained on land-based DAS data does not perform well with submarine DAS due to differences in noise characteristics, deployment conditions and environmental factors. This study presents a machine learning approach tailored specifically to submarine DAS data to enable automated seismic event detection and P- and S-wave identification. Leveraging DeepLab v3, a neural network architecture optimized for semantic segmentation, we developed a specialized model to handle the unique challenges of submarine DAS data. Our model was trained and validated on a data set comprising nearly 57 million manually and semi-automatically labelled seismic records from multiple globally distributed submarine sites, providing a robust basis for accurate seismic detection. The model adapts to a variety of deployment scenarios and can process DAS data from cables with different lengths, configurations and channel spacings, making it versatile for various ocean environments. We thus provide an adaptable and efficient tool for automated earthquake analysis of DAS data, which has the potential to enhance real-time earthquake monitoring and tsunami early warning in submarine environments.
Keywords: 
access_permission:modify_annotations: true
dc:creator: Xiao Han, Tilmann Frederik, van den Ende Martijn, Rivet Diane, Loureiro Afonso, Tsuji Takeshi, Ugalde Arantza, Shi Qibin, Denolle Marine A.
description: DOI: 10.1093/gji/ggag061 Geophysical Journal International, 245, 2, 9-2-2026. Abstract: Given the scarcity of seismometers in marine environments, traditional seismology has limited effectiveness in oceanic regions. Submarine Distributed Acoustic Sensing (DAS) systems offer a promising alternative for seismic monitoring in these areas. However, the existing machine learning model trained on land-based DAS data does not perform well with submarine DAS due to differences in noise characteristics, deployment conditions and environmental factors. This study presents a machine learning approach tailored specifically to submarine DAS data to enable automated seismic event detection and P- and S-wave identification. Leveraging DeepLab v3, a neural network architecture optimized for semantic segmentation, we developed a specialized model to handle the unique challenges of submarine DAS data. Our model was trained and validated on a data set comprising nearly 57 million manually and semi-automatically labelled seismic records from multiple globally distributed submarine sites, providing a robust basis for accurate seismic detection. The model adapts to a variety of deployment scenarios and can process DAS data from cables with different lengths, configurations and channel spacings, making it versatile for various ocean environments. We thus provide an adaptable and efficient tool for automated earthquake analysis of DAS data, which has the potential to enhance real-time earthquake monitoring and tsunami early warning in submarine environments.
dcterms:created: 2026-03-10T09:43:56Z
Last-Modified: 2026-04-10T12:28:57Z
dcterms:modified: 2026-04-10T12:28:57Z
title: DeepSubDAS: an earthquake phase picker from submarine distributed acoustic sensing data
xmpMM:DocumentID: uuid:3d7083fe-d177-3bf3-b857-ea329af1e291
Last-Save-Date: 2026-04-10T12:28:57Z
pdf:docinfo:keywords: 
pdf:docinfo:modified: 2026-04-10T12:28:57Z
meta:save-date: 2026-04-10T12:28:57Z
Content-Type: application/pdf
X-Parsed-By: org.apache.tika.parser.DefaultParser
creator: Xiao Han, Tilmann Frederik, van den Ende Martijn, Rivet Diane, Loureiro Afonso, Tsuji Takeshi, Ugalde Arantza, Shi Qibin, Denolle Marine A.
dc:language: English
dc:subject: 
access_permission:assemble_document: true
xmpTPg:NPages: 16
pdf:charsPerPage: 4650
access_permission:extract_content: true
access_permission:can_print: true
meta:keyword: 
access_permission:can_modify: true
pdf:docinfo:created: 2026-03-10T09:43:56Z