Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology
We introduce a proof of concept to parametrise the unresolved subgrid scale of sea-ice dynamics with deep learning techniques. Instead of parametrising single processes, a single neural network is trained to correct all model variables at the same time. This data-driven approach is applied to a regi...
Published in: | The Cryosphere |
---|---|
Main Authors: | , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
European Geosciences Union
2023
|
Subjects: | |
Online Access: | https://centaur.reading.ac.uk/112714/ https://centaur.reading.ac.uk/112714/1/tc-17-2965-2023.pdf |
id |
ftunivreading:oai:centaur.reading.ac.uk:112714 |
---|---|
record_format |
openpolar |
spelling |
ftunivreading:oai:centaur.reading.ac.uk:112714 2024-06-23T07:56:40+00:00 Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology Finn, Tobias Sebastian Durand, Charlotte Farchi, Alban Bocquet, Marc Chen, Yumeng Carrassi, Alberto Dansereau, Véronique 2023-07 text https://centaur.reading.ac.uk/112714/ https://centaur.reading.ac.uk/112714/1/tc-17-2965-2023.pdf en eng European Geosciences Union https://centaur.reading.ac.uk/112714/1/tc-17-2965-2023.pdf Finn, T. S., Durand, C., Farchi, A., Bocquet, M., Chen, Y. <https://centaur.reading.ac.uk/view/creators/90010659.html> orcid:0000-0002-2319-6937 , Carrassi, A. <https://centaur.reading.ac.uk/view/creators/90010334.html> orcid:0000-0003-0722-5600 and Dansereau, V. (2023) Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology. The Cryosphere, 17 (7). pp. 2965-2991. ISSN 1994-0424 doi: https://doi.org/10.5194/tc-17-2965-2023 <https://doi.org/10.5194/tc-17-2965-2023> cc_by_4 Article PeerReviewed 2023 ftunivreading https://doi.org/10.5194/tc-17-2965-2023 2024-06-11T15:12:32Z We introduce a proof of concept to parametrise the unresolved subgrid scale of sea-ice dynamics with deep learning techniques. Instead of parametrising single processes, a single neural network is trained to correct all model variables at the same time. This data-driven approach is applied to a regional sea-ice model that accounts exclusively for dynamical processes with a Maxwell elasto-brittle rheology. Driven by an external wind forcing in a 40 km×200 km domain, the model generates examples of sharp transitions between unfractured and fully fractured sea ice. To correct such examples, we propose a convolutional U-Net architecture which extracts features at multiple scales. We test this approach in twin experiments: the neural network learns to correct forecasts from low-resolution simulations towards high-resolution simulations for a lead time of about 10 min. At this lead time, our approach reduces the forecast errors by more than 75 %, averaged over all model variables. As the most important predictors, we identify the dynamics of the model variables. Furthermore, the neural network extracts localised and directional-dependent features, which point towards the shortcomings of the low-resolution simulations. Applied to correct the forecasts every 10 min, the neural network is run together with the sea-ice model. This improves the short-term forecasts up to an hour. These results consequently show that neural networks can correct model errors from the subgrid scale for sea-ice dynamics. We therefore see this study as an important first step towards hybrid modelling to forecast sea-ice dynamics on an hourly to daily timescale. Article in Journal/Newspaper Sea ice The Cryosphere CentAUR: Central Archive at the University of Reading The Cryosphere 17 7 2965 2991 |
institution |
Open Polar |
collection |
CentAUR: Central Archive at the University of Reading |
op_collection_id |
ftunivreading |
language |
English |
description |
We introduce a proof of concept to parametrise the unresolved subgrid scale of sea-ice dynamics with deep learning techniques. Instead of parametrising single processes, a single neural network is trained to correct all model variables at the same time. This data-driven approach is applied to a regional sea-ice model that accounts exclusively for dynamical processes with a Maxwell elasto-brittle rheology. Driven by an external wind forcing in a 40 km×200 km domain, the model generates examples of sharp transitions between unfractured and fully fractured sea ice. To correct such examples, we propose a convolutional U-Net architecture which extracts features at multiple scales. We test this approach in twin experiments: the neural network learns to correct forecasts from low-resolution simulations towards high-resolution simulations for a lead time of about 10 min. At this lead time, our approach reduces the forecast errors by more than 75 %, averaged over all model variables. As the most important predictors, we identify the dynamics of the model variables. Furthermore, the neural network extracts localised and directional-dependent features, which point towards the shortcomings of the low-resolution simulations. Applied to correct the forecasts every 10 min, the neural network is run together with the sea-ice model. This improves the short-term forecasts up to an hour. These results consequently show that neural networks can correct model errors from the subgrid scale for sea-ice dynamics. We therefore see this study as an important first step towards hybrid modelling to forecast sea-ice dynamics on an hourly to daily timescale. |
format |
Article in Journal/Newspaper |
author |
Finn, Tobias Sebastian Durand, Charlotte Farchi, Alban Bocquet, Marc Chen, Yumeng Carrassi, Alberto Dansereau, Véronique |
spellingShingle |
Finn, Tobias Sebastian Durand, Charlotte Farchi, Alban Bocquet, Marc Chen, Yumeng Carrassi, Alberto Dansereau, Véronique Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology |
author_facet |
Finn, Tobias Sebastian Durand, Charlotte Farchi, Alban Bocquet, Marc Chen, Yumeng Carrassi, Alberto Dansereau, Véronique |
author_sort |
Finn, Tobias Sebastian |
title |
Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology |
title_short |
Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology |
title_full |
Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology |
title_fullStr |
Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology |
title_full_unstemmed |
Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology |
title_sort |
deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a maxwell elasto-brittle rheology |
publisher |
European Geosciences Union |
publishDate |
2023 |
url |
https://centaur.reading.ac.uk/112714/ https://centaur.reading.ac.uk/112714/1/tc-17-2965-2023.pdf |
genre |
Sea ice The Cryosphere |
genre_facet |
Sea ice The Cryosphere |
op_relation |
https://centaur.reading.ac.uk/112714/1/tc-17-2965-2023.pdf Finn, T. S., Durand, C., Farchi, A., Bocquet, M., Chen, Y. <https://centaur.reading.ac.uk/view/creators/90010659.html> orcid:0000-0002-2319-6937 , Carrassi, A. <https://centaur.reading.ac.uk/view/creators/90010334.html> orcid:0000-0003-0722-5600 and Dansereau, V. (2023) Deep learning subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell elasto-brittle rheology. The Cryosphere, 17 (7). pp. 2965-2991. ISSN 1994-0424 doi: https://doi.org/10.5194/tc-17-2965-2023 <https://doi.org/10.5194/tc-17-2965-2023> |
op_rights |
cc_by_4 |
op_doi |
https://doi.org/10.5194/tc-17-2965-2023 |
container_title |
The Cryosphere |
container_volume |
17 |
container_issue |
7 |
container_start_page |
2965 |
op_container_end_page |
2991 |
_version_ |
1802649935983673344 |