Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions
Subglacial seismicity presents the opportunity to monitor inaccessible glacial beds at the epicentral location and time. Glaciers can be underlain by rock or till, a first order control on bed mechanics. Velocity-weakening, necessary for unstable slip, has been shown for each bed type, but is much s...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Cambridge University Press
|
| Series: | Annals of Glaciology |
| Subjects: | |
| Online Access: | https://www.cambridge.org/core/product/identifier/S0260305524000119/type/journal_article |
| _version_ | 1797218205230432256 |
|---|---|
| author | Seth Saltiel Nathan Groebner Theresa Sawi Christine McCarthy |
| author_facet | Seth Saltiel Nathan Groebner Theresa Sawi Christine McCarthy |
| author_sort | Seth Saltiel |
| collection | DOAJ |
| description | Subglacial seismicity presents the opportunity to monitor inaccessible glacial beds at the epicentral location and time. Glaciers can be underlain by rock or till, a first order control on bed mechanics. Velocity-weakening, necessary for unstable slip, has been shown for each bed type, but is much stronger and evolves over more than an order of magnitude longer distances for till beds. Utilizing a de-stiffened double direct shear apparatus, we found conditions for instability at freezing temperatures and high slip rates for both bed types. During stick–slip stress-drops, we recorded acoustic emissions with piezoelectric transducers frozen into the ice. The two populations of event waveforms appear visually similar and overlap in their statistical features. We implemented a suite of supervised machine learning algorithms to classify the bed type of recorded waveforms and spectra, with prediction accuracy between 65–80%. The Random Forest Classifier is interpretable, showing the importance of initial oscillation peaks and higher frequency energy. Till beds have generally higher friction and resulting stress-drops, with more impulsive first arrivals and more high frequency content compared to rock emissions, but rock beds can produce many till-like events. Seismic signatures could enhance interpretation of bed conditions and mechanics from subglacial seismicity. |
| first_indexed | 2024-04-24T12:14:03Z |
| format | Article |
| id | doaj.art-dc86d0774e744f8e9ea2f5f0ebbdfc6b |
| institution | Directory Open Access Journal |
| issn | 0260-3055 1727-5644 |
| language | English |
| last_indexed | 2024-04-24T12:14:03Z |
| publisher | Cambridge University Press |
| record_format | Article |
| series | Annals of Glaciology |
| spelling | doaj.art-dc86d0774e744f8e9ea2f5f0ebbdfc6b2024-04-08T08:49:59ZengCambridge University PressAnnals of Glaciology0260-30551727-56441810.1017/aog.2024.11Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissionsSeth Saltiel0https://orcid.org/0000-0002-8058-6894Nathan Groebner1Theresa Sawi2Christine McCarthy3Earth and Atmospheric Sciences, Cornell University, Ithaca, NY, USADepartment of Research, Strabo Analytics, Inc, New York, NY, USALamont-Doherty Earth Observatory, Columbia University of New York, NY, USALamont-Doherty Earth Observatory, Columbia University of New York, NY, USASubglacial seismicity presents the opportunity to monitor inaccessible glacial beds at the epicentral location and time. Glaciers can be underlain by rock or till, a first order control on bed mechanics. Velocity-weakening, necessary for unstable slip, has been shown for each bed type, but is much stronger and evolves over more than an order of magnitude longer distances for till beds. Utilizing a de-stiffened double direct shear apparatus, we found conditions for instability at freezing temperatures and high slip rates for both bed types. During stick–slip stress-drops, we recorded acoustic emissions with piezoelectric transducers frozen into the ice. The two populations of event waveforms appear visually similar and overlap in their statistical features. We implemented a suite of supervised machine learning algorithms to classify the bed type of recorded waveforms and spectra, with prediction accuracy between 65–80%. The Random Forest Classifier is interpretable, showing the importance of initial oscillation peaks and higher frequency energy. Till beds have generally higher friction and resulting stress-drops, with more impulsive first arrivals and more high frequency content compared to rock emissions, but rock beds can produce many till-like events. Seismic signatures could enhance interpretation of bed conditions and mechanics from subglacial seismicity.https://www.cambridge.org/core/product/identifier/S0260305524000119/type/journal_articleSeismicitysubglacial exploration geophysicssubglacial processesseismologyglacier geophysics |
| spellingShingle | Seth Saltiel Nathan Groebner Theresa Sawi Christine McCarthy Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions Annals of Glaciology Seismicity subglacial exploration geophysics subglacial processes seismology glacier geophysics |
| title | Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions |
| title_full | Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions |
| title_fullStr | Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions |
| title_full_unstemmed | Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions |
| title_short | Characterization of seismicity from different glacial bed types: machine learning classification of laboratory stick-slip acoustic emissions |
| title_sort | characterization of seismicity from different glacial bed types machine learning classification of laboratory stick slip acoustic emissions |
| topic | Seismicity subglacial exploration geophysics subglacial processes seismology glacier geophysics |
| url | https://www.cambridge.org/core/product/identifier/S0260305524000119/type/journal_article |
| work_keys_str_mv | AT sethsaltiel characterizationofseismicityfromdifferentglacialbedtypesmachinelearningclassificationoflaboratorystickslipacousticemissions AT nathangroebner characterizationofseismicityfromdifferentglacialbedtypesmachinelearningclassificationoflaboratorystickslipacousticemissions AT theresasawi characterizationofseismicityfromdifferentglacialbedtypesmachinelearningclassificationoflaboratorystickslipacousticemissions AT christinemccarthy characterizationofseismicityfromdifferentglacialbedtypesmachinelearningclassificationoflaboratorystickslipacousticemissions |