CaltechAUTHORS
  A Caltech Library Service

Data-Driven Inference of the Mechanics of Slip Along Glacier Beds Using Physics-Informed Neural Networks: Case Study on Rutford Ice Stream, Antarctica

Riel, B. and Minchew, B. and Bischoff, T. (2021) Data-Driven Inference of the Mechanics of Slip Along Glacier Beds Using Physics-Informed Neural Networks: Case Study on Rutford Ice Stream, Antarctica. Journal of Advances in Modeling Earth Systems, 13 (11). Art. No. e2021MS002621. ISSN 1942-2466. doi:10.1029/2021MS002621. https://resolver.caltech.edu/CaltechAUTHORS:20211015-222200700

[img] PDF - Published Version
Creative Commons Attribution.

5MB
[img] PDF - Supplemental Material
Creative Commons Attribution.

3MB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20211015-222200700

Abstract

Reliable projections of sea-level rise depend on accurate representations of how fast-flowing glaciers slip along their beds. The mechanics of slip are often parameterized as a constitutive relation (or “sliding law”) whose proper form remains uncertain. Here, we present a novel deep learning-based framework for learning the time evolution of drag at glacier beds from time-dependent ice velocity and elevation observations. We use a feedforward neural network, informed by the governing equations of ice flow, to infer spatially and temporally varying basal drag and associated uncertainties from data. We test the framework on 1D and 2D ice flow simulation outputs and demonstrate the recovery of the underlying basal mechanics under various levels of observational and modeling uncertainties. We apply this framework to time-dependent velocity data for Rutford Ice Stream, Antarctica, and present evidence that ocean-tide-driven changes in subglacial water pressure drive changes in ice flow over the tidal cycle.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1029/2021ms002621DOIArticle
https://doi.org/10.5281/zenodo.4755695Related Itemsimulation data and deep learning training code
https://doi.org/10.5281/zenodo.5508397Related Itemderived velocity components for Rutford Ice Stream, Antarctica
https://nsidc.org/data/NSIDC-0756/versions/1Related ItemBed topography for Antarctica
ORCID:
AuthorORCID
Riel, B.0000-0003-1940-3910
Minchew, B.0000-0002-5991-3926
Bischoff, T.0000-0003-3930-2762
Additional Information:© 2021 The Authors. Journal of Advances in Modeling Earth Systems published by Wiley Periodicals LLC on behalf of American Geophysical Union. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Issue Online: 10 November 2021; Version of Record online: 10 November 2021; Accepted manuscript online: 23 September 2021; Manuscript accepted: 17 September 2021; Manuscript revised: 15 September 2021; Manuscript received: 12 May 2021. The authors thank three anonymous reviewers for their constructive feedback for improving the quality of this work. Funding for this work was provided by the Earl A Killian III (1978) and Waidy Lee Fund and the NEC Corporation Fund for Research in Computers and Communications. Computing resources were partially funded through a Microsoft AI For Earth computing grant. The authors declare that they have no competing interests. Data Availability Statement: All simulation data and deep learning training code can be downloaded at https://doi.org/10.5281/zenodo.4755695. Derived velocity components for Rutford Ice Stream, Antarctica can be downloaded at https://doi.org/10.5281/zenodo.5508397. Original COSMO-SkyMed products (copyright ASI Agenzia Spaziale Italiana, 2013–2016) are archived at Jet Propulsion Laboratory and were processed under license from ASI as part of a collaborative project between CIDOT and JPL/Caltech. Bed topography for Antarctica available at https://nsidc.org/data/NSIDC-0756/versions/1.
Funders:
Funding AgencyGrant Number
Earl A Killian III (1978) and Waidy Lee FundUNSPECIFIED
NEC CorporationUNSPECIFIED
MicrosoftUNSPECIFIED
Subject Keywords:Glacier dynamics; Physics-informed neural networks; Glacier basal mechanics; Remote sensing; Machine learning
Issue or Number:11
DOI:10.1029/2021MS002621
Record Number:CaltechAUTHORS:20211015-222200700
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20211015-222200700
Official Citation:Riel, B., Minchew, B., & Bischoff, T. (2021). Data-driven inference of the mechanics of slip along glacier beds using physics-informed neural networks: Case study on Rutford Ice Stream, Antarctica. Journal of Advances in Modeling Earth Systems, 13, e2021MS002621. https://doi.org/10.1029/2021MS002621
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:111478
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:18 Oct 2021 15:03
Last Modified:16 Nov 2021 17:13

Repository Staff Only: item control page