Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models
Arctic sea ice prediction is of great practical significance in facilitating Arctic route planning, optimizing fisheries management, and advancing the field of sea ice dynamics research. While various deep learning models have been developed for sea ice prediction, they predominantly operate at the...
Published in: | Journal of Marine Science and Engineering |
---|---|
Main Authors: | , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2023
|
Subjects: | |
Online Access: | https://doi.org/10.3390/jmse11122319 https://doaj.org/article/082241254f524e5aa0f89cdb572902c5 |
id |
ftdoajarticles:oai:doaj.org/article:082241254f524e5aa0f89cdb572902c5 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:082241254f524e5aa0f89cdb572902c5 2024-01-21T10:03:11+01:00 Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models Juanjuan Feng Jia Li Wenjie Zhong Junhui Wu Zhiqiang Li Lingshuai Kong Lei Guo 2023-12-01T00:00:00Z https://doi.org/10.3390/jmse11122319 https://doaj.org/article/082241254f524e5aa0f89cdb572902c5 EN eng MDPI AG https://www.mdpi.com/2077-1312/11/12/2319 https://doaj.org/toc/2077-1312 doi:10.3390/jmse11122319 2077-1312 https://doaj.org/article/082241254f524e5aa0f89cdb572902c5 Journal of Marine Science and Engineering, Vol 11, Iss 12, p 2319 (2023) sea ice concentration recurrent neural network Arctic sea ice prediction short-term prediction Naval architecture. Shipbuilding. Marine engineering VM1-989 Oceanography GC1-1581 article 2023 ftdoajarticles https://doi.org/10.3390/jmse11122319 2023-12-24T01:36:59Z Arctic sea ice prediction is of great practical significance in facilitating Arctic route planning, optimizing fisheries management, and advancing the field of sea ice dynamics research. While various deep learning models have been developed for sea ice prediction, they predominantly operate at the seasonal or sub-seasonal scale, often focusing on localized areas, and few cater to full-region daily-scale prediction. This study introduces the use of spatiotemporal sequence data prediction models, namely, the convolutional LSTM (ConvLSTM) and predictive recurrent neural network (PredRNN), for the prediction of sea ice concentration (SIC). Our analysis reveals that, when solely utilizing SIC historical data as the input, the ConvLSTM model outperforms the PredRNN model in SIC prediction. To enhance the models’ capacity to capture spatiotemporal relationships between multiple variables, we expanded the range of input data types to form the ConvLSTM-multi and PredRNN-multi models. Experimental findings demonstrate that the prediction accuracy of the four models significantly surpasses the CMIP6 model in three prospective climate scenarios (SSP126, SSP245, and SSP585). Of the four models, the ConvLSTM-multi model excels in assimilating the influence of reanalysis data on sea ice within the sea ice edge region, thus exhibiting superior performance than the PredRNN-multi model in predicting daily Arctic SIC over the subsequent 10 days. Furthermore, sensitivity tests on various model parameters highlight the substantial impact of sea surface temperature and prediction date on the accuracy of daily sea ice prediction, and meteorological and oceanographic parameters primarily affect the prediction accuracy of the thin-ice region at the edge of the sea ice. Article in Journal/Newspaper Arctic Sea ice Directory of Open Access Journals: DOAJ Articles Arctic Journal of Marine Science and Engineering 11 12 2319 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
sea ice concentration recurrent neural network Arctic sea ice prediction short-term prediction Naval architecture. Shipbuilding. Marine engineering VM1-989 Oceanography GC1-1581 |
spellingShingle |
sea ice concentration recurrent neural network Arctic sea ice prediction short-term prediction Naval architecture. Shipbuilding. Marine engineering VM1-989 Oceanography GC1-1581 Juanjuan Feng Jia Li Wenjie Zhong Junhui Wu Zhiqiang Li Lingshuai Kong Lei Guo Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models |
topic_facet |
sea ice concentration recurrent neural network Arctic sea ice prediction short-term prediction Naval architecture. Shipbuilding. Marine engineering VM1-989 Oceanography GC1-1581 |
description |
Arctic sea ice prediction is of great practical significance in facilitating Arctic route planning, optimizing fisheries management, and advancing the field of sea ice dynamics research. While various deep learning models have been developed for sea ice prediction, they predominantly operate at the seasonal or sub-seasonal scale, often focusing on localized areas, and few cater to full-region daily-scale prediction. This study introduces the use of spatiotemporal sequence data prediction models, namely, the convolutional LSTM (ConvLSTM) and predictive recurrent neural network (PredRNN), for the prediction of sea ice concentration (SIC). Our analysis reveals that, when solely utilizing SIC historical data as the input, the ConvLSTM model outperforms the PredRNN model in SIC prediction. To enhance the models’ capacity to capture spatiotemporal relationships between multiple variables, we expanded the range of input data types to form the ConvLSTM-multi and PredRNN-multi models. Experimental findings demonstrate that the prediction accuracy of the four models significantly surpasses the CMIP6 model in three prospective climate scenarios (SSP126, SSP245, and SSP585). Of the four models, the ConvLSTM-multi model excels in assimilating the influence of reanalysis data on sea ice within the sea ice edge region, thus exhibiting superior performance than the PredRNN-multi model in predicting daily Arctic SIC over the subsequent 10 days. Furthermore, sensitivity tests on various model parameters highlight the substantial impact of sea surface temperature and prediction date on the accuracy of daily sea ice prediction, and meteorological and oceanographic parameters primarily affect the prediction accuracy of the thin-ice region at the edge of the sea ice. |
format |
Article in Journal/Newspaper |
author |
Juanjuan Feng Jia Li Wenjie Zhong Junhui Wu Zhiqiang Li Lingshuai Kong Lei Guo |
author_facet |
Juanjuan Feng Jia Li Wenjie Zhong Junhui Wu Zhiqiang Li Lingshuai Kong Lei Guo |
author_sort |
Juanjuan Feng |
title |
Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models |
title_short |
Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models |
title_full |
Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models |
title_fullStr |
Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models |
title_full_unstemmed |
Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models |
title_sort |
daily-scale prediction of arctic sea ice concentration based on recurrent neural network models |
publisher |
MDPI AG |
publishDate |
2023 |
url |
https://doi.org/10.3390/jmse11122319 https://doaj.org/article/082241254f524e5aa0f89cdb572902c5 |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic Sea ice |
genre_facet |
Arctic Sea ice |
op_source |
Journal of Marine Science and Engineering, Vol 11, Iss 12, p 2319 (2023) |
op_relation |
https://www.mdpi.com/2077-1312/11/12/2319 https://doaj.org/toc/2077-1312 doi:10.3390/jmse11122319 2077-1312 https://doaj.org/article/082241254f524e5aa0f89cdb572902c5 |
op_doi |
https://doi.org/10.3390/jmse11122319 |
container_title |
Journal of Marine Science and Engineering |
container_volume |
11 |
container_issue |
12 |
container_start_page |
2319 |
_version_ |
1788693439670386688 |