An Unsupervised Machine‐Learning Approach to Understanding Seismicity at an Alpine Glacier

Sawi, Theresa; Holtzman, B. K.; Walter, Fabian; Paisley, John

doi:10.1029/2022jf006909

Cited by 8 publications

(3 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the unsupervised ML method SpecUFEx (Holtzman et al, 2018) to discriminate between various types of seismic events recorded by the Axial OBSs. This spectral feature extraction method was originally developed for audio signal recognition (Cotton and Ellis, 2011) and has been adapted to characterize seismic signals in various settings, such as earthquakes in geothermal elds and along crustal faults, acoustic emissions in lab experiments, and icequakes and seismic noises at glaciers (Holtzman et al, 2018(Holtzman et al, , 2021Sawi et al, 2022). SpecUFEx generates low dimensional spectral ngerprints for each earthquake signal which are then clustered to nd groups of similar signals (Figure 1).…”

Section: Unsupervised ML For Event Discriminationmentioning

confidence: 99%

Real-Time Detection of Volcanic Unrest and Eruption at Axial Seamount using Machine Learning

Wang,

Waldhauser,

Schaff

et al. 2024

Preprint

View full text Add to dashboard Cite

Axial Seamount, an extensively instrumented submarine volcano, lies at the intersection of the Cobb-Eickelberg hot spot and the Juan de Fuca Ridge. Since late 2014, the Ocean Observatories Initiative (OOI) has operated a seven-station cabled ocean bottom seismometers (OBS) array that captured Axial's last eruption in April 2015. This network streams data in real-time, facilitating seismic monitoring and analysis for volcanic unrest detection and eruption forecasting. In this study, we introduce a machine learning (ML) based real-time seismic monitoring framework for Axial Seamount. Combining both supervised and unsupervised ML and double-difference techniques, we constructed a comprehensive, high-resolution earthquake catalog and effectively discriminated between various seismic and acoustic events. These signals include earthquakes generated by different physical processes, acoustic signals of lava-water interaction, and oceanic sources such as whale calls. We first built a labeled ML-based earthquake catalog that extends from November 2014 to the end of 2021, and then implemented real-time monitoring and seismic analysis starting in 2022. With rapid determination of high-resolution earthquake locations and the capability to track potential precursory signals and co-eruption indicators of magma outflow, this system may improve eruption forecasting by providing short-term constraints on Axial's next eruption. Furthermore, our work demonstrated an effective application that integrates unsupervised learning for signal discrimination in real-time operation, which could be generalized to other regions for volcanic unrest detection and enhanced eruption forecasting.

show abstract

Section: Unsupervised ML For Event Discriminationmentioning

confidence: 99%

Real-Time Detection of Volcanic Unrest and Eruption at Axial Seamount using Machine Learning

Wang,

Waldhauser,

Schaff

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The citation impact of articles in the field of glacial lake hazards has exhibited an exponential growth trend since 2000. This indicates, on one hand, that new technologies like remote sensing, GIS, and deep learning have provided more data sources and analytical tools for glacial lake identification research (Lu et al, 2020;Sawi et al, 2022). On the other hand, in recent years, topics such as the interaction between glacial lakes and climate change, glacial lake water resources management, and glacial lake-related disasters have increased in frequency.…”

Section: Data Sourcesmentioning

confidence: 99%

Three decades of glacial lake research: a bibliometric and visual analysis of glacial lake identification

Liu,

Yang,

et al. 2023

Front. Ecol. Evol.

View full text Add to dashboard Cite

IntroductionAs a vital component within glacier systems, the occurrences of glacial lake disasters in high mountain regions have progressively emerged as one of the most destructive natural calamities amid the backdrop of global warming. The swift advancement of glacial lake identification techniques offers a pivotal perspective for forecasting and mitigating the perils of glacial lake outburst disasters.MethodsTo evaluate the thematic evolution, research focal points, and forthcoming directions within the glacier identification domain, a comprehensive bibliometric analysis was conducted on glacial lake identification-related literature from 1991 to 2023 in the Web of Science Core Collection database.Results1) The United States, the United Kingdom, and China stand as principal nations propelling the field's advancement. The Chinese Academy of Sciences demonstrates the highest activity in terms of article publications and international collaborations. 2) Climate change, compilation of glacial lake inventories, methodologies for risk assessment, glacial lake outburst floods, comprehensive disaster management strategies, and hydrodynamic models constitute the domain's research hotspots. It is a typical multidisciplinary field. 3) Persistently high-impact topics over an extended period include “hazard”, “Late Pleistocene”, “environmental change”, “ice sheet”, and “lake sediments”. Keywords indicating the present cutting-edge research encompass “inventory”, “glacial lake outburst flood”, “risk”, “dynamics”, “Tibetan Plateau”, “evolution”, and “high mountain Asia”.DiscussionThis paper delves into the current status and pivotal concerns of glacial lake identification techniques, methodologies, and the scale of identification research themes. Further Research should concentrate on avenues like “recognition methods grounded in machine learning and deep learning”, “multisource data fusion datasets”, “novel algorithms and technologies adaptable to scale transformation and data expansion”, as well as “enhancing spatiotemporal data resolution”. This will ultimately enable precise, prolonged, and multiscalar identification of glacial lakes. his study provides valuable guidance and reference for future research in the field of glacial lake identification.

show abstract

“…For example, automated detection and classification of seismic events has been used to identify an ice-shelf fracture process induced by tidal bending, followed by resonance as seawater presumably fills the new fracture (Hammer et al, 2015). By studying the timing of icequake rupture, englacial meltwater flow has been related to other processes, such as heating by solar radiation and cooling induced by katabatic winds (Helmstetter et al, 2015a;Sawi et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Towards the systematic reconnaissance of seismic signals from glaciers and ice sheets – Part B: Unsupervised learning for source process characterisation

Latto¹,

Turner²,

Reading³

et al. 2023

Preprint

View full text Add to dashboard Cite

Abstract. Given the high number and diversity of events in a typical cryoseismic dataset, in particular those recorded on ice sheet margins, it is desirable to use a semi-automated method of grouping similar events for reconnaissance and ongoing analysis. We present a workflow for employing semi-unsupervised cluster analysis to inform investigations of the processes occurring in glaciers and ice sheets. In this demonstration study, we make use of a seismic event catalogue previously compiled for the Whillans Ice Stream, for the 2010–2011 austral summer (outlined in companion paper, Latto et al., 2023). We address the challenges of seismic event analysis for a complex wavefield by clustering similar seismic events into groups using characteristic temporal, spectral, and polarization attributes of seismic time series with the k-means++ algorithm. This provides the basis for a reconnaissance analysis of a seismic wavefield that contains local events (from the ice stream) set in an ambient wavefield that itself contains a diversity of signals (mostly from the Ross Ice Shelf). As one result, we find that two clusters include stick-slip events that diverge in terms of length and initiation locality (i.e. Central Sticky Spot and/or the grounding line). We also identify a swarm of high frequency signals on January 16–17, 2011 that are potentially associated with a surface melt event from the Ross Ice Shelf. Used together with the event detection presented in the companion paper, the semi-automated workflow could readily generalize to other locations, and as a possible benchmark procedure, could enable the monitoring of remote glaciers over time and comparisons between locations.

show abstract

An Unsupervised Machine‐Learning Approach to Understanding Seismicity at an Alpine Glacier

Cited by 8 publications

References 64 publications

Real-Time Detection of Volcanic Unrest and Eruption at Axial Seamount using Machine Learning

Real-Time Detection of Volcanic Unrest and Eruption at Axial Seamount using Machine Learning

Three decades of glacial lake research: a bibliometric and visual analysis of glacial lake identification

Towards the systematic reconnaissance of seismic signals from glaciers and ice sheets – Part B: Unsupervised learning for source process characterisation

Contact Info

Product

Resources

About