Attention Network Forecasts Time‐to‐Failure in Laboratory Shear Experiments

Jasperson, Hope; Bolton, David; Johnson, P. A.; Guyer, R. A.; Marone, Chris; Hoop, Maarten V. de

doi:10.1029/2021jb022195

Cited by 22 publications

(19 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The application of machine learning using continuous seismic records continues to show success in describing physical processes of complex natural systems. While the glacier motion model predictions are not as robust as those for laboratory stick-slip studies (Corbi et al, 2019;Jasperson et al, 2021;Rouet-LeDuc, Hulbert, Bolton, et al, 2018;Shokouhi et al, 2021;Wang et al, 2021), slow slip in Earth (Hulbert et al, 2020), future prediction (Laurenti et al, 2022;Wang et al, 2022), or stick-slip processes in Earth , they are nonetheless predictive for the long-term sliding behavior and especially performant for short-term variations. Ice deformation is considered mostly aseismic through viscous creep (Gimbert et al, 2021), which is inherent to the material properties.…”

Section: Discussionmentioning

confidence: 86%

Mapping Glacier Basal Sliding Applying Machine Learning

Umlauft,

Johnson,

Roux

et al. 2023

JGR Earth Surface

Self Cite

View full text Add to dashboard Cite

During the RESOLVE project (“High‐resolution imaging in subsurface geophysics: development of a multi‐instrument platform for interdisciplinary research”), continuous surface displacement and seismic array observations were obtained on Glacier d’Argentière in the French Alps for 35 days in May 2018. The data set is used to perform a detailed study of targeted processes within the highly dynamic cryospheric environment. In particular, the physical processes controlling glacial basal motion are poorly understood and remain challenging to observe directly. Especially in the Alpine region for temperate based glaciers where the ice rapidly responds to changing climatic conditions and thus, processes are strongly intermittent in time and heterogeneous in space. Spatially dense seismic and Global Positioning System (GPS) measurements are analyzed applying machine learning to gain insight into the processes controlling glacial motions of Glacier d’Argentière. Using multiple bandpass‐filtered copies of the continuous seismic waveforms, we compute energy‐based features, develop a matched field beamforming catalog and include meteorological observations. Features describing the data are analyzed with a gradient boosting decision tree model to directly estimate the GPS displacements from the seismic noise. We posit that features of the seismic noise provide direct access to the dominant parameters that drive displacement on the highly variable and unsteady surface of the glacier. The machine learning model infers daily fluctuations and longer term trends. The results show on‐ice displacement rates are strongly modulated by activity at the base of the glacier. The techniques presented provide a new approach to study glacial basal sliding and discover its full complexity.

show abstract

Section: Discussionmentioning

confidence: 86%

Mapping Glacier Basal Sliding Applying Machine Learning

Umlauft,

Johnson,

Roux

et al. 2023

JGR Earth Surface

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 90%

Mapping glacier basal sliding with machine learning

Umlauft¹,

Johnson²,

Roux³

et al. 2023

Preprint

View full text Add to dashboard Cite

During the RESOLVE project ("High-resolution imaging in subsurface geophysics: development of a multi-instrument platform for interdisciplinary research"), continuous surface displacement and seismic array observations were obtained on Glacier d'Argentière in the French Alps for 35 days during May in 2018. This unique data set offers the chance to perform a detailed, local study of targeted processes within the highly dynamic cryospheric environment. In particular, the physical processes controlling glacial basal motion are poorly understood and remain challenging to observe directly. Especially in the Alpine region for temperate based glaciers where the ice rapidly responds to changing climatic conditions and thus, processes are strongly intermittent in time and heterogeneous in space. Spatially dense seismic and GPS measurements are analyzed with machine learning techniques to gain insight into the underlying processes controlling glacial motions of Glacier d'Argentière.Using multiple bandpass-filtered copies of the continuous seismic waveforms, we compute energy-based features, develop a matched field beamforming catalogue and include meteorological observations. Features describing the data are analyzed with a gradient boosting decision tree model to directly estimate the GPS displacements from the seismic records.We posit that features of the seismic noise provide direct access to the dominant parameters that drive displacement on the highly variable and unsteady surface of the glacier. The machine learning model infers daily fluctuations as well as longer term trends and the results show on-ice displacement rates are strongly modulated by activity at the base of the glacier. The techniques presented provide a new approach to study glacial basal sliding and discover its full complexity.

show abstract

“…By focusing the attention of the neural networks on specific features, fracture loading mode (Z. and fracture saturation are successfully inferred from the laboratory earthquakes. Similarly, although the deterministic prediction of time-to-failure in natural environments remains elusive, a couple of studies (Jasperson et al, 2021;Shreedharan et al, 2021) show that ML has the ability to predict time-to-failure and the stress state from the laboratory earthquake data. J.-T. Lin et al (2021) demonstrates that deep neural networks trained by simulated ground deformation data have the potential to provide accurate early warnings for large earthquakes, particularly when the regional tectonic setting is well understood and data are abundant.…”

Section: Earthquake Data Applicationsmentioning

confidence: 99%

“…Song et al., 2022) and fracture saturation (Nolte & Pyrak‐Nolte, 2022) are successfully inferred from the laboratory earthquakes. Similarly, although the deterministic prediction of time‐to‐failure in natural environments remains elusive, a couple of studies (Jasperson et al., 2021; Shreedharan et al., 2021) show that ML has the ability to predict time‐to‐failure and the stress state from the laboratory earthquake data. J.‐T.…”

Section: Highlightsmentioning

confidence: 99%

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

O’Malley

Beroza

et al. 2023

JGR Solid Earth

Self Cite

View full text Add to dashboard Cite

After decades of low but continuing activity, applications of machine learning (ML) in solid Earth geoscience have exploded in popularity. This special collection provides a snapshot of those applications, which range from data processing to inversion and interpretation, for which ML appears particularly well suited. Inevitably, there are variations in the degree to which these methods have been developed. We hope that the progress seen in some areas will inspire efforts in others. Challenges remain, including the formidable task of how geoscience can keep pace with developments in ML while ensuring the scientific rigor that our field depends on, but with improvements in sensor technology and accelerating rates of data accumulation, the methods of ML seem poised to play an important role for the foreseeable future.

show abstract

Attention Network Forecasts Time‐to‐Failure in Laboratory Shear Experiments

Cited by 22 publications

References 75 publications

Mapping Glacier Basal Sliding Applying Machine Learning

Mapping Glacier Basal Sliding Applying Machine Learning

Mapping glacier basal sliding with machine learning

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

Contact Info

Product

Resources

About