Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings
El Niño-Southern Oscillation (ENSO), a cyclic climate phenomenon spanning interannual and decadal timescales, exerts substantial impacts on the global weather patterns and ecosystems. Recently, deep learning has brought considerable advances in the accurate prediction of ENSO occurrence. However, th...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-02-01
|
| Series: | Frontiers in Marine Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmars.2024.1334210/full |
| _version_ | 1797316773756796928 |
|---|---|
| author | Chuang Rui Zhengya Sun Zhengya Sun Wensheng Zhang An-An Liu Zhiqiang Wei |
| author_facet | Chuang Rui Zhengya Sun Zhengya Sun Wensheng Zhang An-An Liu Zhiqiang Wei |
| author_sort | Chuang Rui |
| collection | DOAJ |
| description | El Niño-Southern Oscillation (ENSO), a cyclic climate phenomenon spanning interannual and decadal timescales, exerts substantial impacts on the global weather patterns and ecosystems. Recently, deep learning has brought considerable advances in the accurate prediction of ENSO occurrence. However, the current models are insufficient to characterize the evolutionary behavior of the ENSO, particularly lacking comprehensive modeling of local-range and longrange spatiotemporal interdependencies, and the incorporation of calendar monthly and seasonal properties. To make up this gap, we propose a Two-Stage SpatioTemporal (TSST) autoregressive model that couples the meteorological factor prediction with ENSO indicator prediction. The first stage predicts the meteorological time series by leveraging self-attention ConvLSTM network which captures both the local and the global spatial-temporal dependencies. The temporal embeddings of calendar months and seasonal information are further incorporated to preserves repeatedly-occurring-yet-hidden patterns in meteorological series. The second stage uses multiple layers to extract higher level of features from predicted meteorological factors progressively to generate ENSO indicators. The results demonstrate that our model outperforms the state-of-the-art ENSO prediction models, effectively predicting ENSO up to 24 months and mitigating the spring predictability barrier. |
| first_indexed | 2024-03-08T03:24:13Z |
| format | Article |
| id | doaj.art-2c0279e3f28147c78ad6b439c974808f |
| institution | Directory Open Access Journal |
| issn | 2296-7745 |
| language | English |
| last_indexed | 2024-03-08T03:24:13Z |
| publishDate | 2024-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Marine Science |
| spelling | doaj.art-2c0279e3f28147c78ad6b439c974808f2024-02-12T04:43:55ZengFrontiers Media S.A.Frontiers in Marine Science2296-77452024-02-011110.3389/fmars.2024.13342101334210Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddingsChuang Rui0Zhengya Sun1Zhengya Sun2Wensheng Zhang3An-An Liu4Zhiqiang Wei5State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute ofAutomation, Chinese Academy of Sciences, Beijing, ChinaState Key Laboratory of Multimodal Artificial Intelligence Systems, Institute ofAutomation, Chinese Academy of Sciences, Beijing, ChinaSchool of Artificial Intelligence, University of Chinese Academy of Sciences (UCAS), Beijing, ChinaState Key Laboratory of Multimodal Artificial Intelligence Systems, Institute ofAutomation, Chinese Academy of Sciences, Beijing, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaDepartment of Computer Science and Technology, Ocean University of China, Qingdao, Shandong, ChinaEl Niño-Southern Oscillation (ENSO), a cyclic climate phenomenon spanning interannual and decadal timescales, exerts substantial impacts on the global weather patterns and ecosystems. Recently, deep learning has brought considerable advances in the accurate prediction of ENSO occurrence. However, the current models are insufficient to characterize the evolutionary behavior of the ENSO, particularly lacking comprehensive modeling of local-range and longrange spatiotemporal interdependencies, and the incorporation of calendar monthly and seasonal properties. To make up this gap, we propose a Two-Stage SpatioTemporal (TSST) autoregressive model that couples the meteorological factor prediction with ENSO indicator prediction. The first stage predicts the meteorological time series by leveraging self-attention ConvLSTM network which captures both the local and the global spatial-temporal dependencies. The temporal embeddings of calendar months and seasonal information are further incorporated to preserves repeatedly-occurring-yet-hidden patterns in meteorological series. The second stage uses multiple layers to extract higher level of features from predicted meteorological factors progressively to generate ENSO indicators. The results demonstrate that our model outperforms the state-of-the-art ENSO prediction models, effectively predicting ENSO up to 24 months and mitigating the spring predictability barrier.https://www.frontiersin.org/articles/10.3389/fmars.2024.1334210/fullEl Niño-Southern Oscillation (ENSO)deep learning for ENSO predictionself-attention ConvLSTMtemporal embeddingsspring prediction barrier |
| spellingShingle | Chuang Rui Zhengya Sun Zhengya Sun Wensheng Zhang An-An Liu Zhiqiang Wei Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings Frontiers in Marine Science El Niño-Southern Oscillation (ENSO) deep learning for ENSO prediction self-attention ConvLSTM temporal embeddings spring prediction barrier |
| title | Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings |
| title_full | Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings |
| title_fullStr | Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings |
| title_full_unstemmed | Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings |
| title_short | Enhancing ENSO predictions with self-attention ConvLSTM and temporal embeddings |
| title_sort | enhancing enso predictions with self attention convlstm and temporal embeddings |
| topic | El Niño-Southern Oscillation (ENSO) deep learning for ENSO prediction self-attention ConvLSTM temporal embeddings spring prediction barrier |
| url | https://www.frontiersin.org/articles/10.3389/fmars.2024.1334210/full |
| work_keys_str_mv | AT chuangrui enhancingensopredictionswithselfattentionconvlstmandtemporalembeddings AT zhengyasun enhancingensopredictionswithselfattentionconvlstmandtemporalembeddings AT zhengyasun enhancingensopredictionswithselfattentionconvlstmandtemporalembeddings AT wenshengzhang enhancingensopredictionswithselfattentionconvlstmandtemporalembeddings AT ananliu enhancingensopredictionswithselfattentionconvlstmandtemporalembeddings AT zhiqiangwei enhancingensopredictionswithselfattentionconvlstmandtemporalembeddings |