Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations
Abstract In this study, we perform online sea ice bias correction within a Geophysical Fluid Dynamics Laboratory global ice‐ocean model. For this, we use a convolutional neural network (CNN) which was developed in a previous study (Gregory et al., 2023, https://doi.org/10.1029/2023ms003757) for the...
Published in: | Geophysical Research Letters |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2024
|
Subjects: | |
Online Access: | https://doi.org/10.1029/2023GL106776 https://doaj.org/article/c1f2f3c4aff14bcaac00ef9cebfba463 |
id |
ftdoajarticles:oai:doaj.org/article:c1f2f3c4aff14bcaac00ef9cebfba463 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:c1f2f3c4aff14bcaac00ef9cebfba463 2024-09-09T20:06:48+00:00 Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations William Gregory Mitchell Bushuk Yongfei Zhang Alistair Adcroft Laure Zanna 2024-02-01T00:00:00Z https://doi.org/10.1029/2023GL106776 https://doaj.org/article/c1f2f3c4aff14bcaac00ef9cebfba463 EN eng Wiley https://doi.org/10.1029/2023GL106776 https://doaj.org/toc/0094-8276 https://doaj.org/toc/1944-8007 1944-8007 0094-8276 doi:10.1029/2023GL106776 https://doaj.org/article/c1f2f3c4aff14bcaac00ef9cebfba463 Geophysical Research Letters, Vol 51, Iss 3, Pp n/a-n/a (2024) sea ice machine learning data assimilation modeling parameterization neural networks Geophysics. Cosmic physics QC801-809 article 2024 ftdoajarticles https://doi.org/10.1029/2023GL106776 2024-08-05T17:49:23Z Abstract In this study, we perform online sea ice bias correction within a Geophysical Fluid Dynamics Laboratory global ice‐ocean model. For this, we use a convolutional neural network (CNN) which was developed in a previous study (Gregory et al., 2023, https://doi.org/10.1029/2023ms003757) for the purpose of predicting sea ice concentration (SIC) data assimilation (DA) increments. An initial implementation of the CNN shows systematic improvements in SIC biases relative to the free‐running model, however large summertime errors remain. We show that these residual errors can be significantly improved with a novel sea ice data augmentation approach. This approach applies sequential CNN and DA corrections to a new simulation over the training period, which then provides a new training data set to refine the weights of the initial network. We propose that this machine‐learned correction scheme could be utilized for generating improved initial conditions, and also for real‐time sea ice bias correction within seasonal‐to‐subseasonal sea ice forecasts. Article in Journal/Newspaper Sea ice Directory of Open Access Journals: DOAJ Articles Geophysical Research Letters 51 3 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
sea ice machine learning data assimilation modeling parameterization neural networks Geophysics. Cosmic physics QC801-809 |
spellingShingle |
sea ice machine learning data assimilation modeling parameterization neural networks Geophysics. Cosmic physics QC801-809 William Gregory Mitchell Bushuk Yongfei Zhang Alistair Adcroft Laure Zanna Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations |
topic_facet |
sea ice machine learning data assimilation modeling parameterization neural networks Geophysics. Cosmic physics QC801-809 |
description |
Abstract In this study, we perform online sea ice bias correction within a Geophysical Fluid Dynamics Laboratory global ice‐ocean model. For this, we use a convolutional neural network (CNN) which was developed in a previous study (Gregory et al., 2023, https://doi.org/10.1029/2023ms003757) for the purpose of predicting sea ice concentration (SIC) data assimilation (DA) increments. An initial implementation of the CNN shows systematic improvements in SIC biases relative to the free‐running model, however large summertime errors remain. We show that these residual errors can be significantly improved with a novel sea ice data augmentation approach. This approach applies sequential CNN and DA corrections to a new simulation over the training period, which then provides a new training data set to refine the weights of the initial network. We propose that this machine‐learned correction scheme could be utilized for generating improved initial conditions, and also for real‐time sea ice bias correction within seasonal‐to‐subseasonal sea ice forecasts. |
format |
Article in Journal/Newspaper |
author |
William Gregory Mitchell Bushuk Yongfei Zhang Alistair Adcroft Laure Zanna |
author_facet |
William Gregory Mitchell Bushuk Yongfei Zhang Alistair Adcroft Laure Zanna |
author_sort |
William Gregory |
title |
Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations |
title_short |
Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations |
title_full |
Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations |
title_fullStr |
Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations |
title_full_unstemmed |
Machine Learning for Online Sea Ice Bias Correction Within Global Ice‐Ocean Simulations |
title_sort |
machine learning for online sea ice bias correction within global ice‐ocean simulations |
publisher |
Wiley |
publishDate |
2024 |
url |
https://doi.org/10.1029/2023GL106776 https://doaj.org/article/c1f2f3c4aff14bcaac00ef9cebfba463 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
Geophysical Research Letters, Vol 51, Iss 3, Pp n/a-n/a (2024) |
op_relation |
https://doi.org/10.1029/2023GL106776 https://doaj.org/toc/0094-8276 https://doaj.org/toc/1944-8007 1944-8007 0094-8276 doi:10.1029/2023GL106776 https://doaj.org/article/c1f2f3c4aff14bcaac00ef9cebfba463 |
op_doi |
https://doi.org/10.1029/2023GL106776 |
container_title |
Geophysical Research Letters |
container_volume |
51 |
container_issue |
3 |
_version_ |
1809939346742050816 |