OxyGenerator: Reconstructing Global Ocean Deoxygenation Over a Century with Deep Learning

Bin Lu, Ze Zhao, Luyu Han, Xiaoying Gan, Yuntao Zhou, Lei Zhou, Luoyi Fu, Xinbing Wang, Chenghu Zhou, Jing Zhang
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:33219-33242, 2024.

Abstract

Accurately reconstructing the global ocean deoxygenation over a century is crucial for assessing and protecting marine ecosystem. Existing expert-dominated numerical simulations fail to catch up with the dynamic variation caused by global warming and human activities. Besides, due to the high-cost data collection, the historical observations are severely sparse, leading to big challenge for precise reconstruction. In this work, we propose OxyGenerator, the first deep learning based model, to reconstruct the global ocean deoxygenation from 1920 to 2023. Specifically, to address the heterogeneity across large temporal and spatial scales, we propose zoning-varying graph message-passing to capture the complex oceanographic correlations between missing values and sparse observations. Additionally, to further calibrate the uncertainty, we incorporate inductive bias from dissolved oxygen (DO) variations and chemical effects. Compared with in-situ DO observations, OxyGenerator significantly outperforms CMIP6 numerical simulations, reducing MAPE by 38.77%, demonstrating a promising potential to understand the “breathless ocean” in data-driven manner.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-lu24n, title = {{O}xy{G}enerator: Reconstructing Global Ocean Deoxygenation Over a Century with Deep Learning}, author = {Lu, Bin and Zhao, Ze and Han, Luyu and Gan, Xiaoying and Zhou, Yuntao and Zhou, Lei and Fu, Luoyi and Wang, Xinbing and Zhou, Chenghu and Zhang, Jing}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {33219--33242}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/lu24n/lu24n.pdf}, url = {https://proceedings.mlr.press/v235/lu24n.html}, abstract = {Accurately reconstructing the global ocean deoxygenation over a century is crucial for assessing and protecting marine ecosystem. Existing expert-dominated numerical simulations fail to catch up with the dynamic variation caused by global warming and human activities. Besides, due to the high-cost data collection, the historical observations are severely sparse, leading to big challenge for precise reconstruction. In this work, we propose OxyGenerator, the first deep learning based model, to reconstruct the global ocean deoxygenation from 1920 to 2023. Specifically, to address the heterogeneity across large temporal and spatial scales, we propose zoning-varying graph message-passing to capture the complex oceanographic correlations between missing values and sparse observations. Additionally, to further calibrate the uncertainty, we incorporate inductive bias from dissolved oxygen (DO) variations and chemical effects. Compared with in-situ DO observations, OxyGenerator significantly outperforms CMIP6 numerical simulations, reducing MAPE by 38.77%, demonstrating a promising potential to understand the “breathless ocean” in data-driven manner.} }
Endnote
%0 Conference Paper %T OxyGenerator: Reconstructing Global Ocean Deoxygenation Over a Century with Deep Learning %A Bin Lu %A Ze Zhao %A Luyu Han %A Xiaoying Gan %A Yuntao Zhou %A Lei Zhou %A Luoyi Fu %A Xinbing Wang %A Chenghu Zhou %A Jing Zhang %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-lu24n %I PMLR %P 33219--33242 %U https://proceedings.mlr.press/v235/lu24n.html %V 235 %X Accurately reconstructing the global ocean deoxygenation over a century is crucial for assessing and protecting marine ecosystem. Existing expert-dominated numerical simulations fail to catch up with the dynamic variation caused by global warming and human activities. Besides, due to the high-cost data collection, the historical observations are severely sparse, leading to big challenge for precise reconstruction. In this work, we propose OxyGenerator, the first deep learning based model, to reconstruct the global ocean deoxygenation from 1920 to 2023. Specifically, to address the heterogeneity across large temporal and spatial scales, we propose zoning-varying graph message-passing to capture the complex oceanographic correlations between missing values and sparse observations. Additionally, to further calibrate the uncertainty, we incorporate inductive bias from dissolved oxygen (DO) variations and chemical effects. Compared with in-situ DO observations, OxyGenerator significantly outperforms CMIP6 numerical simulations, reducing MAPE by 38.77%, demonstrating a promising potential to understand the “breathless ocean” in data-driven manner.
APA
Lu, B., Zhao, Z., Han, L., Gan, X., Zhou, Y., Zhou, L., Fu, L., Wang, X., Zhou, C. & Zhang, J.. (2024). OxyGenerator: Reconstructing Global Ocean Deoxygenation Over a Century with Deep Learning. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:33219-33242 Available from https://proceedings.mlr.press/v235/lu24n.html.

Related Material