Deep learning of 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...
Main Authors: | , , , , , , |
---|---|
Format: | Text |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://doi.org/10.5194/egusphere-2022-1342 https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1342/ |
id |
ftcopernicus:oai:publications.copernicus.org:egusphere108016 |
---|---|
record_format |
openpolar |
spelling |
ftcopernicus:oai:publications.copernicus.org:egusphere108016 2023-08-15T12:43:01+02:00 Deep learning of 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 2023-07-21 application/pdf https://doi.org/10.5194/egusphere-2022-1342 https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1342/ eng eng doi:10.5194/egusphere-2022-1342 https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1342/ eISSN: Text 2023 ftcopernicus https://doi.org/10.5194/egusphere-2022-1342 2023-07-24T16:24:18Z 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. Text Sea ice Copernicus Publications: E-Journals |
institution |
Open Polar |
collection |
Copernicus Publications: E-Journals |
op_collection_id |
ftcopernicus |
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 |
Text |
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 of 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 of subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell-Elasto-Brittle rheology |
title_short |
Deep learning of subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell-Elasto-Brittle rheology |
title_full |
Deep learning of subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell-Elasto-Brittle rheology |
title_fullStr |
Deep learning of subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell-Elasto-Brittle rheology |
title_full_unstemmed |
Deep learning of subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a Maxwell-Elasto-Brittle rheology |
title_sort |
deep learning of subgrid-scale parametrisations for short-term forecasting of sea-ice dynamics with a maxwell-elasto-brittle rheology |
publishDate |
2023 |
url |
https://doi.org/10.5194/egusphere-2022-1342 https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1342/ |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
eISSN: |
op_relation |
doi:10.5194/egusphere-2022-1342 https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1342/ |
op_doi |
https://doi.org/10.5194/egusphere-2022-1342 |
_version_ |
1774298525557850112 |