A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI
Abstract 25 thousand years ago, the European Alps were covered by the kilometre-thick Alpine Ice Field. Numerical modelling of this glaciation has been challenged by model-data disagreements, including overestimations of ice thickness. We tackle this issue by applying the Instructed Glacier Model, a...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-56168-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585571999940608 |
|---|---|
| author | Tancrède P. M. Leger Guillaume Jouvet Sarah Kamleitner Jürgen Mey Frédéric Herman Brandon D. Finley Susan Ivy-Ochs Andreas Vieli Andreas Henz Samuel U. Nussbaumer |
| author_facet | Tancrède P. M. Leger Guillaume Jouvet Sarah Kamleitner Jürgen Mey Frédéric Herman Brandon D. Finley Susan Ivy-Ochs Andreas Vieli Andreas Henz Samuel U. Nussbaumer |
| author_sort | Tancrède P. M. Leger |
| collection | DOAJ |
| description | Abstract 25 thousand years ago, the European Alps were covered by the kilometre-thick Alpine Ice Field. Numerical modelling of this glaciation has been challenged by model-data disagreements, including overestimations of ice thickness. We tackle this issue by applying the Instructed Glacier Model, a three-dimensional model enhanced with physics-informed machine learning. This approach allows us to produce 100 Alps-wide and 17 thousand-year-long simulations at 300 m resolution. Previously unfeasible due to computational costs, our experiment both increases model-data agreement in ice extent and reduces the offset in ice thickness by between 200% and 450% relative to previous studies. Our results have implications for better estimating former ice velocities, ice temperature, basal conditions, erosion processes, and paleoclimate in the Alps. This study demonstrates that physics-informed machine learning can help overcome the bottleneck of high-resolution glacier modelling and better test parameterisations, both of which are required to accurately describe complex topographies and ice dynamics. |
| format | Article |
| id | doaj-art-6fca2d370e1f4991b8b65d8e641e53e0 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-6fca2d370e1f4991b8b65d8e641e53e02025-01-26T12:40:32ZengNature PortfolioNature Communications2041-17232025-01-0116111610.1038/s41467-025-56168-3A data-consistent model of the last glaciation in the Alps achieved with physics-driven AITancrède P. M. Leger0Guillaume Jouvet1Sarah Kamleitner2Jürgen Mey3Frédéric Herman4Brandon D. Finley5Susan Ivy-Ochs6Andreas Vieli7Andreas Henz8Samuel U. Nussbaumer9Institute of Earth Surface Dynamics, University of LausanneInstitute of Earth Surface Dynamics, University of LausanneInstitute of Earth Surface Dynamics, University of LausanneInstitute of Environmental Science and Geography, University of PotsdamInstitute of Earth Surface Dynamics, University of LausanneInstitute of Earth Surface Dynamics, University of LausanneLaboratory of Ion Beam Physics, ETH ZurichDepartment of Geography, University of ZurichDepartment of Geography, University of ZurichDepartment of Geography, University of ZurichAbstract 25 thousand years ago, the European Alps were covered by the kilometre-thick Alpine Ice Field. Numerical modelling of this glaciation has been challenged by model-data disagreements, including overestimations of ice thickness. We tackle this issue by applying the Instructed Glacier Model, a three-dimensional model enhanced with physics-informed machine learning. This approach allows us to produce 100 Alps-wide and 17 thousand-year-long simulations at 300 m resolution. Previously unfeasible due to computational costs, our experiment both increases model-data agreement in ice extent and reduces the offset in ice thickness by between 200% and 450% relative to previous studies. Our results have implications for better estimating former ice velocities, ice temperature, basal conditions, erosion processes, and paleoclimate in the Alps. This study demonstrates that physics-informed machine learning can help overcome the bottleneck of high-resolution glacier modelling and better test parameterisations, both of which are required to accurately describe complex topographies and ice dynamics.https://doi.org/10.1038/s41467-025-56168-3 |
| spellingShingle | Tancrède P. M. Leger Guillaume Jouvet Sarah Kamleitner Jürgen Mey Frédéric Herman Brandon D. Finley Susan Ivy-Ochs Andreas Vieli Andreas Henz Samuel U. Nussbaumer A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI Nature Communications |
| title | A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI |
| title_full | A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI |
| title_fullStr | A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI |
| title_full_unstemmed | A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI |
| title_short | A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI |
| title_sort | data consistent model of the last glaciation in the alps achieved with physics driven ai |
| url | https://doi.org/10.1038/s41467-025-56168-3 |
| work_keys_str_mv | AT tancredepmleger adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT guillaumejouvet adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT sarahkamleitner adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT jurgenmey adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT fredericherman adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT brandondfinley adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT susanivyochs adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT andreasvieli adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT andreashenz adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT samuelunussbaumer adataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT tancredepmleger dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT guillaumejouvet dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT sarahkamleitner dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT jurgenmey dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT fredericherman dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT brandondfinley dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT susanivyochs dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT andreasvieli dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT andreashenz dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai AT samuelunussbaumer dataconsistentmodelofthelastglaciationinthealpsachievedwithphysicsdrivenai |