Generalizable deep learning models for predicting laboratory earthquakes

Abstract Machine learning models can predict laboratory earthquakes using Acoustic emission, the lab equivalent of microseismicity, and changes in fault zone elastic properties during the lab seismic cycle. Applying them to natural earthquakes requires testing their generalizability across lab setti...

Full description

Saved in:
Bibliographic Details
Main Authors: Chonglang Wang, Kaiwen Xia, Wei Yao, Chris Marone
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Communications Earth & Environment
Online Access:https://doi.org/10.1038/s43247-025-02200-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849389986118369280
author Chonglang Wang
Kaiwen Xia
Wei Yao
Chris Marone
author_facet Chonglang Wang
Kaiwen Xia
Wei Yao
Chris Marone
author_sort Chonglang Wang
collection DOAJ
description Abstract Machine learning models can predict laboratory earthquakes using Acoustic emission, the lab equivalent of microseismicity, and changes in fault zone elastic properties during the lab seismic cycle. Applying them to natural earthquakes requires testing their generalizability across lab settings and stress conditions. Here, we show a fine-tuned convolutional neural network (CNN) model effectively transfer across different conditions. Our model employs techniques from natural language processing, including decoder techniques, to capture the relationship between AE and fault stress. We fine-tune the regression head of a deep CNN while fixing the decoder’s weights and successfully predict lab seismic events for a range of conditions. With fine-tuning, CNN models trained on one lab fault configuration predict time to failure and shear stress for another configuration at varying fault slip rates. These results demonstrate the potential of extending lab-based methods to different conditions that could eventually include tectonic earthquakes and seismic forecasting.
format Article
id doaj-art-b8b3ae23f26e4bcbb629576cd485b3d2
institution Kabale University
issn 2662-4435
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Communications Earth & Environment
spelling doaj-art-b8b3ae23f26e4bcbb629576cd485b3d22025-08-20T03:41:47ZengNature PortfolioCommunications Earth & Environment2662-44352025-03-016111010.1038/s43247-025-02200-9Generalizable deep learning models for predicting laboratory earthquakesChonglang Wang0Kaiwen Xia1Wei Yao2Chris Marone3State Key Laboratory of Hydraulic Engineering Intelligent Construction and Operation, School of Civil Engineering, Tianjin UniversityState Key Laboratory of Hydraulic Engineering Intelligent Construction and Operation, School of Civil Engineering, Tianjin UniversityState Key Laboratory of Hydraulic Engineering Intelligent Construction and Operation, School of Civil Engineering, Tianjin UniversityDipartimento di Scienze della Terra, La Sapienza Università di RomaAbstract Machine learning models can predict laboratory earthquakes using Acoustic emission, the lab equivalent of microseismicity, and changes in fault zone elastic properties during the lab seismic cycle. Applying them to natural earthquakes requires testing their generalizability across lab settings and stress conditions. Here, we show a fine-tuned convolutional neural network (CNN) model effectively transfer across different conditions. Our model employs techniques from natural language processing, including decoder techniques, to capture the relationship between AE and fault stress. We fine-tune the regression head of a deep CNN while fixing the decoder’s weights and successfully predict lab seismic events for a range of conditions. With fine-tuning, CNN models trained on one lab fault configuration predict time to failure and shear stress for another configuration at varying fault slip rates. These results demonstrate the potential of extending lab-based methods to different conditions that could eventually include tectonic earthquakes and seismic forecasting.https://doi.org/10.1038/s43247-025-02200-9
spellingShingle Chonglang Wang
Kaiwen Xia
Wei Yao
Chris Marone
Generalizable deep learning models for predicting laboratory earthquakes
Communications Earth & Environment
title Generalizable deep learning models for predicting laboratory earthquakes
title_full Generalizable deep learning models for predicting laboratory earthquakes
title_fullStr Generalizable deep learning models for predicting laboratory earthquakes
title_full_unstemmed Generalizable deep learning models for predicting laboratory earthquakes
title_short Generalizable deep learning models for predicting laboratory earthquakes
title_sort generalizable deep learning models for predicting laboratory earthquakes
url https://doi.org/10.1038/s43247-025-02200-9
work_keys_str_mv AT chonglangwang generalizabledeeplearningmodelsforpredictinglaboratoryearthquakes
AT kaiwenxia generalizabledeeplearningmodelsforpredictinglaboratoryearthquakes
AT weiyao generalizabledeeplearningmodelsforpredictinglaboratoryearthquakes
AT chrismarone generalizabledeeplearningmodelsforpredictinglaboratoryearthquakes