Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning

Abstract Effective use of the wealth of information provided by Distributed Acoustic Sensing (DAS) for mass movement monitoring remains a challenge. We propose a semi‐supervised neural network tailored to screen DAS data related to a series of rock collapses leading to a major failure of approximate...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiahui Kang, Fabian Walter, Patrick Paitz, Johannes Aichele, Pascal Edme, Lorenz Meier, Andreas Fichtner
Format: Article
Language:English
Published: Wiley 2024-10-01
Series:Geophysical Research Letters
Subjects:
Online Access:https://doi.org/10.1029/2024GL110672
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850229037424705536
author Jiahui Kang
Fabian Walter
Patrick Paitz
Johannes Aichele
Pascal Edme
Lorenz Meier
Andreas Fichtner
author_facet Jiahui Kang
Fabian Walter
Patrick Paitz
Johannes Aichele
Pascal Edme
Lorenz Meier
Andreas Fichtner
author_sort Jiahui Kang
collection DOAJ
description Abstract Effective use of the wealth of information provided by Distributed Acoustic Sensing (DAS) for mass movement monitoring remains a challenge. We propose a semi‐supervised neural network tailored to screen DAS data related to a series of rock collapses leading to a major failure of approximately 1.2 million m3 on 15 June 2023 in Brienz, Eastern Switzerland. Besides DAS, the dataset from 16 May to 30 June 2023 includes Doppler radar data for partially ground‐truth labeling. The proposed algorithm is capable of distinguishing between rock‐slope failures and background noise, including road and train traffic, with a detection precision of over 95%. It identifies hundreds of precursory failures and shows sustained detection hours before and during the major collapse. Event size and signal‐to‐noise ratio (SNR) are the key performance dependencies. As a critical part of our algorithm operates unsupervised, we suggest that it is suitable for general monitoring of natural hazards.
format Article
id doaj-art-bfb09a042b7b44cea68452cdd127aea6
institution OA Journals
issn 0094-8276
1944-8007
language English
publishDate 2024-10-01
publisher Wiley
record_format Article
series Geophysical Research Letters
spelling doaj-art-bfb09a042b7b44cea68452cdd127aea62025-08-20T02:04:21ZengWileyGeophysical Research Letters0094-82761944-80072024-10-015119n/an/a10.1029/2024GL110672Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised LearningJiahui Kang0Fabian Walter1Patrick Paitz2Johannes Aichele3Pascal Edme4Lorenz Meier5Andreas Fichtner6Swiss Federal Institute for Forest Snow and Landscape Research Zürich SwitzerlandSwiss Federal Institute for Forest Snow and Landscape Research Zürich SwitzerlandSwiss Federal Institute for Forest Snow and Landscape Research Zürich SwitzerlandDepartment of Earth Sciences ETH Zürich Zürich SwitzerlandDepartment of Earth Sciences ETH Zürich Zürich SwitzerlandGeopraevent AG Zürich SwitzerlandDepartment of Earth Sciences ETH Zürich Zürich SwitzerlandAbstract Effective use of the wealth of information provided by Distributed Acoustic Sensing (DAS) for mass movement monitoring remains a challenge. We propose a semi‐supervised neural network tailored to screen DAS data related to a series of rock collapses leading to a major failure of approximately 1.2 million m3 on 15 June 2023 in Brienz, Eastern Switzerland. Besides DAS, the dataset from 16 May to 30 June 2023 includes Doppler radar data for partially ground‐truth labeling. The proposed algorithm is capable of distinguishing between rock‐slope failures and background noise, including road and train traffic, with a detection precision of over 95%. It identifies hundreds of precursory failures and shows sustained detection hours before and during the major collapse. Event size and signal‐to‐noise ratio (SNR) are the key performance dependencies. As a critical part of our algorithm operates unsupervised, we suggest that it is suitable for general monitoring of natural hazards.https://doi.org/10.1029/2024GL110672distributed acoustic sensingmachine learningprecursorsimage processingrepresentation learningearly warning
spellingShingle Jiahui Kang
Fabian Walter
Patrick Paitz
Johannes Aichele
Pascal Edme
Lorenz Meier
Andreas Fichtner
Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning
Geophysical Research Letters
distributed acoustic sensing
machine learning
precursors
image processing
representation learning
early warning
title Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning
title_full Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning
title_fullStr Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning
title_full_unstemmed Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning
title_short Automatic Monitoring of Rock‐Slope Failures Using Distributed Acoustic Sensing and Semi‐Supervised Learning
title_sort automatic monitoring of rock slope failures using distributed acoustic sensing and semi supervised learning
topic distributed acoustic sensing
machine learning
precursors
image processing
representation learning
early warning
url https://doi.org/10.1029/2024GL110672
work_keys_str_mv AT jiahuikang automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning
AT fabianwalter automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning
AT patrickpaitz automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning
AT johannesaichele automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning
AT pascaledme automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning
AT lorenzmeier automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning
AT andreasfichtner automaticmonitoringofrockslopefailuresusingdistributedacousticsensingandsemisupervisedlearning