Interpretable Adaptive Multiscale Spatiotemporal Network for Long-Term Global Sea Surface Temperature Prediction.
Saved in:
| Title: | Interpretable Adaptive Multiscale Spatiotemporal Network for Long-Term Global Sea Surface Temperature Prediction. |
|---|---|
| Authors: | Hao, Rixu1,2 (AUTHOR), Zhao, Yuxin1,2 (AUTHOR) zhaoyuxin@hrbeu.edu.cn, Deng, Xiong1,2 (AUTHOR) |
| Source: | Remote Sensing. Apr2026, Vol. 18 Issue 7, p997. 36p. |
| Subjects: | Ocean dynamics, Wavelet transforms, Ocean temperature, Long short-term memory, Climate change, Machine learning |
| Abstract: | Highlights: What are the main findings? We propose PAMSTnet, a physics-informed adaptive multiscale spatiotemporal network for long-term global SST prediction that unifies multiscale ocean dynamics learning, cross-variable mode representation, and physics-consistency enforcement with thermodynamic and dynamic constraints. PAMSTnet leverages three-dimensional empirical wavelet transform (3DEWT) for interpretable multiscale dynamics learning, multivariate spatiotemporal empirical orthogonal function (MSTEOF) for cross-variable coupled mode representation, and deep ConvLSTM predictive network (DCPN) for coordinated multiscale evolution modeling, and physics-informed consistency learning (PICL) to enforce thermodynamic and dynamic constraints. What are the implications of the main findings? Experiments on global SST predictions demonstrate substantially improved long-term accuracy and robustness to extreme climate variability compared with existing deep learning models, while providing interpretable attribution insights into global SST evolution through physically informed multiscale representations. These findings position PAMSTnet as a promising paradigm for intelligent ocean prediction systems that bridge data-driven multiscale representation learning with physics-consistent dynamical modeling, enabling interpretable long-term ocean predictions. Sea surface temperature (SST) serves as a fundamental driver of ocean–atmosphere interactions and global climate variability, exhibiting strong nonstationarity, multiscale dynamics, and cross-variable coupling. However, current deep learning models often fail to capture these complex characteristics, limiting their ability to support accurate and physically consistent long-term SST prediction. To address these issues, we propose PAMSTnet, a unified deep learning framework for physics-informed adaptive multiscale spatiotemporal prediction. PAMSTnet leverages three-dimensional empirical wavelet transform (3DEWT) to learn interpretable multiscale spatiotemporal dynamics from raw observations, and applies multivariate spatiotemporal empirical orthogonal function (MSTEOF) to identify dominant cross-variable coupled modes. These physically meaningful representations are integrated into a deep ConvLSTM predictive network (DCPN) to support coordinated multiscale dynamical learning. Furthermore, PAMSTnet introduces physics-informed consistency learning (PICL) to enforce thermodynamic and dynamic constraints, enhancing physical consistency and interpretability. Extensive experiments demonstrate that PAMSTnet achieves superior performance against state-of-the-art baselines in long-term global SST prediction, reducing RMSE by 8.1% and improving ACC by 2.8% compared with the best-performing baseline, particularly under extreme climate events. Interpretation insights further highlight PAMSTnet's adaptive representation of variable contributions and regional physical drivers. These findings position PAMSTnet as a promising paradigm for developing intelligent ocean prediction systems with enhanced physical consistency and interpretability. [ABSTRACT FROM AUTHOR] |
| Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Highlights: What are the main findings? We propose PAMSTnet, a physics-informed adaptive multiscale spatiotemporal network for long-term global SST prediction that unifies multiscale ocean dynamics learning, cross-variable mode representation, and physics-consistency enforcement with thermodynamic and dynamic constraints. PAMSTnet leverages three-dimensional empirical wavelet transform (3DEWT) for interpretable multiscale dynamics learning, multivariate spatiotemporal empirical orthogonal function (MSTEOF) for cross-variable coupled mode representation, and deep ConvLSTM predictive network (DCPN) for coordinated multiscale evolution modeling, and physics-informed consistency learning (PICL) to enforce thermodynamic and dynamic constraints. What are the implications of the main findings? Experiments on global SST predictions demonstrate substantially improved long-term accuracy and robustness to extreme climate variability compared with existing deep learning models, while providing interpretable attribution insights into global SST evolution through physically informed multiscale representations. These findings position PAMSTnet as a promising paradigm for intelligent ocean prediction systems that bridge data-driven multiscale representation learning with physics-consistent dynamical modeling, enabling interpretable long-term ocean predictions. Sea surface temperature (SST) serves as a fundamental driver of ocean–atmosphere interactions and global climate variability, exhibiting strong nonstationarity, multiscale dynamics, and cross-variable coupling. However, current deep learning models often fail to capture these complex characteristics, limiting their ability to support accurate and physically consistent long-term SST prediction. To address these issues, we propose PAMSTnet, a unified deep learning framework for physics-informed adaptive multiscale spatiotemporal prediction. PAMSTnet leverages three-dimensional empirical wavelet transform (3DEWT) to learn interpretable multiscale spatiotemporal dynamics from raw observations, and applies multivariate spatiotemporal empirical orthogonal function (MSTEOF) to identify dominant cross-variable coupled modes. These physically meaningful representations are integrated into a deep ConvLSTM predictive network (DCPN) to support coordinated multiscale dynamical learning. Furthermore, PAMSTnet introduces physics-informed consistency learning (PICL) to enforce thermodynamic and dynamic constraints, enhancing physical consistency and interpretability. Extensive experiments demonstrate that PAMSTnet achieves superior performance against state-of-the-art baselines in long-term global SST prediction, reducing RMSE by 8.1% and improving ACC by 2.8% compared with the best-performing baseline, particularly under extreme climate events. Interpretation insights further highlight PAMSTnet's adaptive representation of variable contributions and regional physical drivers. These findings position PAMSTnet as a promising paradigm for developing intelligent ocean prediction systems with enhanced physical consistency and interpretability. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 20724292 |
| DOI: | 10.3390/rs18070997 |
