Aug 2026· Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2· 0 citations· 20 references
TL;DR
The first oceanic 4D sparse observation reconstruction dataset, named OceanVerse, is presented, providing a novel large-scale dataset that meets the MNAR (Missing Not at Random) condition, supporting more effective model comparison, generalization evaluation and potential advancement of scientific reconstruction architectures.
Abstract
The vast oceans record the impacts of climate change and human activities on the Earth system. Over the past century, oceanographic scientists have collected extensive ocean profile data to reflect variations of oceanic elements, such as dissolved oxygen. However, due to the sophisticated measurements and high costs, historical ocean element observation data remains highly sparse and uneven across the global ocean, with the annual missing rate exceeding 90%. Thus, quantitatively understanding the four-dimensional (4D) spatiotemporal evolution of oceanic elements continues to pose a significant challenge. Machine learning (ML) techniques demonstrate superior capabilities in perceiving spatiotemporal variations within large-scale data, presenting promising opportunities to harness implicit correlations for global reconstruction. However, fragmented data and interdisciplinary differences create barriers to the availability of AI-ready open data, further hindering ML practitioners from designing specialized models. To solve this problem, we present the first oceanic 4D sparse observation reconstruction dataset, named OceanVerse. By integrating nearly 2 million real-world profiles since 1900 and three differentiated Earth system numerical simulation, we construct a comprehensively evaluable dataset with missing patterns that align with real-world conditions through a digital twin sampling. OceanVerse provides a novel large-scale (~100× nodes vs. existing datasets) dataset that meets the MNAR (Missing Not at Random) condition, supporting more effective model comparison, generalization evaluation and potential advancement of scientific reconstruction architectures. The OceanVerse dataset and codebase are publicly available. The OceanVerse resource is available at https://github.com/jingwei-sjtu/OceanVerse.
OceanDepths is introduced, the first open, global, regridded AI-ready dataset that pairs satellite-derived sea surface temperature, sea surface salinity, and sea surface height products with co-located EN4 subsurface temperature and salinity profiles, complemented by matched GLORYS12 ocean reanalysis data to support comparisons or multi-stage learning.
Simon Donike, Ruben Cartuyvels, A. I. Ferola et al.· 0 citations
Jingwei is a 4D spatiotemporal graph transfer learning model that reconstructs ocean oxygen levels globally over the past six decades on a 1◦ x 1◦ grid, covering depths of 0-5500 meters, and foresee Jingwei revolutionizing observation-based ocean modeling and deepening the authors' understanding of the breathless ocean.
Bin Lu, Luyu Han, Ze Zhao et al.· Proceedings of the 32nd ACM...· 0 citations
OceanBench is a benchmark designed to evaluate and accelerate global short-range data-driven ocean forecasting, constructed from a curated dataset comprising first-guess trajectories, nowcasts, and atmospheric forcings from operational physical ocean models, typically unavailable in public datasets due to assimilation cycles.
Anass El, Quentin Gaudel, Juan Emmanuel Johnson et al.· Advances in Neural Informati...· 7 citations· ⚡2
4DVarGen is proposed, a 4DVar-inspired generative framework for reconstructing sea surface variable fields at eddy-resolving scales from sparse remote-sensing observations that establishes a mathematical equivalence between 4DVar and an observation-guided denoising process.
Junpeng Huang, Wuxin Wang, Xiaoyong Li et al.· 0 citations
OceanLight achieves pointwise forecast accuracy and kinetic energy spectral fidelity exceeding both operational numerical analyses and state-of-the-art AI-based models, while surpassing all AI-based ocean models in geostrophic balance consistency.
High-precision reconstruction of subsurface ocean temperatures is of great significance for the identification and monitoring of mesoscale phenomena and the prediction of climate change trends. Although advanced artificial intelligent (AI) algorithms have been widely applied to reconstruct subsurface thermohaline fields from surface inputs, most of them cannot be trained on in situ data because of the sparsity and discontinuity of Argo profiles. Considering Argo profiles as labels, only one-dimensional AI algorithms are applicable thus far, and they cannot fully capture the spatiotemporal evolutionary characteristics of the ocean. As a basic property of mesoscale to larger-scale oceans, quasi-geostrophic (QG) dynamics can help estimate the interior parameters of the ocean. Combining QG with AI algorithms is promising for improving the performance of subsurface reconstruction, which has not been fully exploited by current studies. In this study, a hybrid dynamic-statistical reconstruction framework named interior+surface quasi-geostrophic (isQG)-empowered ResNet is proposed, which is based on the one-dimensional residual network (1D-ResNet) and incorporates the density anomalies reconstructed by the isQG method. The results demonstrate the following. First, the model performance is better than that of the conventional 1D-ResNet model, with the layer-averaged root mean square error being reduced by approximately 0.18 $^{\circ }$C. Shapley additive explanations analysis verifies the important contribution of isQG factors. Second, the predicted temperature profiles are in good agreement with the Argo in situ profiles, reasonably reflecting the vertical temperature variations in actual marine conditions. Third, the reconstructed 3-D temperature field can reasonably represent mesoscale phenomena (e.g., mesoscale eddies) and is in line with the GLORYS12V1 reanalysis data. It is suggested that the proposed method is a feasible and stable framework for subsurface reconstruction and is, thus, helpful for analyzing the three-dimensional structures of mesoscale phenomena.
Xiting Sun, T. Xie, Hengqian Yan et al.· IEEE Journal of Selected Top...· 0 citations