OXYGENERATOR: Reconstructing Global Ocean Deoxygenation Over a Century with Deep Learning
This provides a data-driven solution for assessing marine ecosystem health, addressing limitations of expert-dominated numerical simulations in oceanography.
The paper tackles the problem of reconstructing global ocean deoxygenation over a century (1920-2023) by proposing OxyGenerator, a deep learning model that addresses sparse historical observations and dynamic variations, reducing MAPE by 38.77% compared to CMIP6 numerical simulations.
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.