SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer

Abstract Seismic phase picking is a fundamental task in seismology that is crucial for event detection and earthquake cataloging; however, manual analysis is impractical given the scale of modern seismic networks. We present SegPhase, a novel seismic arrival time picking model designed to efficientl...

Full description

Saved in:
Bibliographic Details
Main Authors: Shinya Katoh, Yoshihisa Iio, Hiromichi Nagao, Hiroshi Katao, Masayo Sawada, Kazuhide Tomisaka
Format: Article
Language:English
Published: SpringerOpen 2025-07-01
Series:Earth, Planets and Space
Subjects:
Online Access:https://doi.org/10.1186/s40623-025-02249-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849767005105684480
author Shinya Katoh
Yoshihisa Iio
Hiromichi Nagao
Hiroshi Katao
Masayo Sawada
Kazuhide Tomisaka
author_facet Shinya Katoh
Yoshihisa Iio
Hiromichi Nagao
Hiroshi Katao
Masayo Sawada
Kazuhide Tomisaka
author_sort Shinya Katoh
collection DOAJ
description Abstract Seismic phase picking is a fundamental task in seismology that is crucial for event detection and earthquake cataloging; however, manual analysis is impractical given the scale of modern seismic networks. We present SegPhase, a novel seismic arrival time picking model designed to efficiently process large-scale seismic data recorded by dense seismic networks in Japan. In contrast to conventional convolution-based models, SegPhase employs a hierarchical vision transformer structure that utilizes multi-head self-attention to dynamically focus on important waveform features, such as P- and S-wave onsets, noise, and coda waves. Compared to PhaseNet, the most widely used deep learning model, SegPhase improved arrival time match rates by ~ 11% and detected ~ 15% more events in continuous waveform tests, particularly enhancing the detection of small-magnitude events. Benchmark evaluations demonstrated that SegPhase achieved high classification performance in identifying P- and S-waves. We also examined the threshold of the output probability values when applying SegPhase to continuous waveforms for which the optimal threshold was unknown. By lowering the threshold to 0.1, we observed an increase in the number of detected events without noticeable changes in the hypocenter location error and observed–calculated discrepancies. This was achieved by more effectively utilizing high-probability picks, which further improved phase association. Based on these results, we recommend a threshold of 0.1 to enhance event detection while maintaining accurate arrival times. Our findings demonstrate that SegPhase enables robust arrival picking across diverse datasets and supports high-resolution seismic monitoring. Graphical Abstract
format Article
id doaj-art-a0a8b9c3129e408ca5ccc765b6871dcc
institution DOAJ
issn 1880-5981
language English
publishDate 2025-07-01
publisher SpringerOpen
record_format Article
series Earth, Planets and Space
spelling doaj-art-a0a8b9c3129e408ca5ccc765b6871dcc2025-08-20T03:04:22ZengSpringerOpenEarth, Planets and Space1880-59812025-07-0177112810.1186/s40623-025-02249-ySegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformerShinya Katoh0Yoshihisa Iio1Hiromichi Nagao2Hiroshi Katao3Masayo Sawada4Kazuhide Tomisaka5Earthquake Research Institute, The University of TokyoDisaster Prevention Research Institute, Kyoto UniversityThe University of TokyoDisaster Prevention Research Institute, Kyoto UniversityDisaster Prevention Research Institute, Kyoto UniversityDisaster Prevention Research Institute, Kyoto UniversityAbstract Seismic phase picking is a fundamental task in seismology that is crucial for event detection and earthquake cataloging; however, manual analysis is impractical given the scale of modern seismic networks. We present SegPhase, a novel seismic arrival time picking model designed to efficiently process large-scale seismic data recorded by dense seismic networks in Japan. In contrast to conventional convolution-based models, SegPhase employs a hierarchical vision transformer structure that utilizes multi-head self-attention to dynamically focus on important waveform features, such as P- and S-wave onsets, noise, and coda waves. Compared to PhaseNet, the most widely used deep learning model, SegPhase improved arrival time match rates by ~ 11% and detected ~ 15% more events in continuous waveform tests, particularly enhancing the detection of small-magnitude events. Benchmark evaluations demonstrated that SegPhase achieved high classification performance in identifying P- and S-waves. We also examined the threshold of the output probability values when applying SegPhase to continuous waveforms for which the optimal threshold was unknown. By lowering the threshold to 0.1, we observed an increase in the number of detected events without noticeable changes in the hypocenter location error and observed–calculated discrepancies. This was achieved by more effectively utilizing high-probability picks, which further improved phase association. Based on these results, we recommend a threshold of 0.1 to enhance event detection while maintaining accurate arrival times. Our findings demonstrate that SegPhase enables robust arrival picking across diverse datasets and supports high-resolution seismic monitoring. Graphical Abstracthttps://doi.org/10.1186/s40623-025-02249-yDeep learningPhase pickingEarthquake detectionManten network
spellingShingle Shinya Katoh
Yoshihisa Iio
Hiromichi Nagao
Hiroshi Katao
Masayo Sawada
Kazuhide Tomisaka
SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer
Earth, Planets and Space
Deep learning
Phase picking
Earthquake detection
Manten network
title SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer
title_full SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer
title_fullStr SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer
title_full_unstemmed SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer
title_short SegPhase: development of arrival time picking models for Japan’s seismic network using the hierarchical vision transformer
title_sort segphase development of arrival time picking models for japan s seismic network using the hierarchical vision transformer
topic Deep learning
Phase picking
Earthquake detection
Manten network
url https://doi.org/10.1186/s40623-025-02249-y
work_keys_str_mv AT shinyakatoh segphasedevelopmentofarrivaltimepickingmodelsforjapansseismicnetworkusingthehierarchicalvisiontransformer
AT yoshihisaiio segphasedevelopmentofarrivaltimepickingmodelsforjapansseismicnetworkusingthehierarchicalvisiontransformer
AT hiromichinagao segphasedevelopmentofarrivaltimepickingmodelsforjapansseismicnetworkusingthehierarchicalvisiontransformer
AT hiroshikatao segphasedevelopmentofarrivaltimepickingmodelsforjapansseismicnetworkusingthehierarchicalvisiontransformer
AT masayosawada segphasedevelopmentofarrivaltimepickingmodelsforjapansseismicnetworkusingthehierarchicalvisiontransformer
AT kazuhidetomisaka segphasedevelopmentofarrivaltimepickingmodelsforjapansseismicnetworkusingthehierarchicalvisiontransformer