Ice-flow model emulator based on physics-informed deep learning

Convolutional neural networks (CNN) trained from high-order ice-flow model realisations have proven to be outstanding emulators in terms of fidelity and computational performance. However, the dependence on an ensemble of realisations of an instructor model renders this strategy difficult to general...

Full description

Saved in:
Bibliographic Details
Main Authors: Guillaume Jouvet, Guillaume Cordonnier
Format: Article
Language:English
Published: Cambridge University Press 2023-12-01
Series:Journal of Glaciology
Subjects:
Online Access:https://www.cambridge.org/core/product/identifier/S0022143023000734/type/journal_article
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846127735143923712
author Guillaume Jouvet
Guillaume Cordonnier
author_facet Guillaume Jouvet
Guillaume Cordonnier
author_sort Guillaume Jouvet
collection DOAJ
description Convolutional neural networks (CNN) trained from high-order ice-flow model realisations have proven to be outstanding emulators in terms of fidelity and computational performance. However, the dependence on an ensemble of realisations of an instructor model renders this strategy difficult to generalise to a variety of ice-flow regimes found in the nature. To overcome this issue, we adopt the approach of physics-informed deep learning, which fuses traditional numerical solutions by finite differences/elements and deep-learning approaches. Here, we train a CNN to minimise the energy associated with high-order ice-flow equations within the time iterations of a glacier evolution model. As a result, our emulator is a promising alternative to traditional solvers thanks to its high computational efficiency (especially on GPU), its high fidelity to the original model, its simplified training (without requiring any data), its capability to handle a variety of ice-flow regimes and memorise previous solutions, and its relatively simple implementation. Embedded into the ‘Instructed Glacier Model’ (IGM) framework, the potential of the emulator is illustrated with three applications including a large-scale high-resolution (2400x4000) forward glacier evolution model, an inverse modelling case for data assimilation, and an ice shelf.
format Article
id doaj-art-4a5a316da97843df8b6d9bd5b32a17c6
institution Kabale University
issn 0022-1430
1727-5652
language English
publishDate 2023-12-01
publisher Cambridge University Press
record_format Article
series Journal of Glaciology
spelling doaj-art-4a5a316da97843df8b6d9bd5b32a17c62024-12-11T10:15:40ZengCambridge University PressJournal of Glaciology0022-14301727-56522023-12-01691941195510.1017/jog.2023.73Ice-flow model emulator based on physics-informed deep learningGuillaume Jouvet0https://orcid.org/0000-0002-8546-8459Guillaume Cordonnier1https://orcid.org/0000-0003-0124-0180Université de Lausanne, IDYST, 1015 Lausanne, SwitzerlandInria, Université Côte d'Azur, Sophia-Antipolis, FranceConvolutional neural networks (CNN) trained from high-order ice-flow model realisations have proven to be outstanding emulators in terms of fidelity and computational performance. However, the dependence on an ensemble of realisations of an instructor model renders this strategy difficult to generalise to a variety of ice-flow regimes found in the nature. To overcome this issue, we adopt the approach of physics-informed deep learning, which fuses traditional numerical solutions by finite differences/elements and deep-learning approaches. Here, we train a CNN to minimise the energy associated with high-order ice-flow equations within the time iterations of a glacier evolution model. As a result, our emulator is a promising alternative to traditional solvers thanks to its high computational efficiency (especially on GPU), its high fidelity to the original model, its simplified training (without requiring any data), its capability to handle a variety of ice-flow regimes and memorise previous solutions, and its relatively simple implementation. Embedded into the ‘Instructed Glacier Model’ (IGM) framework, the potential of the emulator is illustrated with three applications including a large-scale high-resolution (2400x4000) forward glacier evolution model, an inverse modelling case for data assimilation, and an ice shelf.https://www.cambridge.org/core/product/identifier/S0022143023000734/type/journal_articleglacier flowglacier modellingglacier mechanicsice-sheet modelling
spellingShingle Guillaume Jouvet
Guillaume Cordonnier
Ice-flow model emulator based on physics-informed deep learning
Journal of Glaciology
glacier flow
glacier modelling
glacier mechanics
ice-sheet modelling
title Ice-flow model emulator based on physics-informed deep learning
title_full Ice-flow model emulator based on physics-informed deep learning
title_fullStr Ice-flow model emulator based on physics-informed deep learning
title_full_unstemmed Ice-flow model emulator based on physics-informed deep learning
title_short Ice-flow model emulator based on physics-informed deep learning
title_sort ice flow model emulator based on physics informed deep learning
topic glacier flow
glacier modelling
glacier mechanics
ice-sheet modelling
url https://www.cambridge.org/core/product/identifier/S0022143023000734/type/journal_article
work_keys_str_mv AT guillaumejouvet iceflowmodelemulatorbasedonphysicsinformeddeeplearning
AT guillaumecordonnier iceflowmodelemulatorbasedonphysicsinformeddeeplearning