ObsSea4Clim Training #4 AI-Driven Quality Control and Uncertainty Quantification in Essential Ocean Variables

ABOUT THE TRAINING

Duration: 60 minutes

Contents:

As AI is expanding into all areas, this workshop describes AI-based quality control tools and applications, and how such an approach could lead to better quantification of uncertainties within the framework of Essential Ocean Variables and Essential Climate Variables (EOV/ECV). After introducing the theoretical framework (key concepts and scientific background), the participants could learn how AI can help automate and improve data quality checks, leading to more robust EOV datasets. Since the uncertainty is an integral part of an EOV, the role of AI-based quality control approaches is discussed in quantitatively assessing it. Using sea surface temperature (SST) as an example of EOV, participants will also see recent results and gain access to Jupyter Notebooks for some hands-on experimentation.

Audience: Early career researchers and scientists working in oceanography, ocean observing, climate prediction, ESM climate modelling, data management, and the wider climate science scientific community.

Your trainer

Thania Papapostolou is a physical oceanographer, working at the Hellenic Centre for Marine Research with the Operational Oceanography group. Her work revolves around ocean observations, and through various projects, she is involved in data curation for the HCMR’s POSEIDON observatories, quality control of Copernicus Marine In-Situ observations for the Mediterranean region, developing AI-based quality control tools, and exploring observing capabilities that emerge from optical fibre sensing through submarine telecommunication cables.

The Hellenic Centre for Marine Research (HCMR) is Greece’s national research organisation for oceanography and marine sciences, supervised by the General Secretariat for Research and Technology. HCMR conducts research and innovation studies, aiming to protect the hydrosphere, support informed decision-making, and provide products and services that benefit society and the economy, while fostering European and international collaborations.

Subscribe to the upcoming training sessions:

https://www.eventbrite.dk/o/obssea4clim-113344569711 

Reading Resources / References to the presentation:

• Bonino, Giulia, Giuliano Galimberti, Simona Masina, Ronan McAdam, and Emanuela Clementi. “Machine Learning Methods to Predict Sea Surface Temperature and Marine Heatwave Occurrence: A Case Study of the Mediterranean Sea.” Ocean Science 20, no. 2 (March 22, 2024): 417–32. https://doi.org/10.5194/os-20-417-2024.
• Boufeniza, Redouane Larbi, Luo Jingjia, Kemal Adem Abdela, Karam Alsafadi, and Mohammad M Alsahli. “Deep Learning for Sea Surface Temperature Applications: A Comprehensive Bibliometric Analysis and Methodological Approach.” Geo: Geography and Environment 11, no. 2 (2024): e00151. https://doi.org/10.1002/geo2.151.
• Castelão, G. P. “A Machine Learning Approach to Quality Control Oceanographic Data.” Computers and Geosciences 155 (October 1, 2021). https://doi.org/10.1016/j.cageo.2021.104803.
• Castelão, Guilherme. “A Framework to Quality Control Oceanographic Data.” Journal of Open Source Software 5, no. 48 (April 7, 2020): 2063. https://doi.org/10.21105/joss.02063.
• Chaudhary, Lalita, Shakti Sharma, and Mohit Sajwan. “Systematic Literature Review of Various Neural Network Techniques for Sea Surface Temperature Prediction Using Remote Sensing Data.” Archives of Computational Methods in Engineering 30, no. 8 (November 1, 2023): 5071–5103. https://doi.org/10.1007/s11831-023-09970-5.
• Gao, G., B. Yao, Z. Li, D. Duan, and X. Zhang. “Forecasting of Sea Surface Temperature in Eastern Tropical Pacifi c by a Hybrid Multiscale Spatial-Temporal Model Combining Error Correction Map.” IEEE Transactions on Geoscience and Remote Sensing 62 (2024): 1–22. https://doi.org/10.1109/TGRS.2024.3353288.
• Gao, Z., Z. Li, J. Yu, and L. Xu. “Global Spatiotemporal Graph Attention Network for Sea Surface Temperature Prediction.” IEEE Geoscience and Remote Sensing Letters 20 (2023). https://doi.org/10.1109/LGRS.2023.3250237.
• Good, Simon A., Matthew J. Martin, and Nick A. Rayner. “EN4: Quality Controlled Ocean Temperature and Salinity Profi les and Monthly Objective Analyses with Uncertainty Estimates.” Journal of Geophysical Research: Oceans 118, no. 12 (2013): 6704–16. https://doi.org/10.1002/2013JC009067.
• Cowley R, Killick RE, Boyer T, Gouretski V, Reseghetti F, Kizu S, Palmer MD, Cheng L, Storto A, Le Menn M, Simoncelli S, Macdonald AM and Domingues CM (2021) International Quality-Controlled Ocean Database (IQuOD) v0.1: The Temperature Uncertainty Specifi cation. Front. Mar. Sci. 8:689695, https://doi.org/10.3389/fmars.2021.689695
• Good, S. A., M. J. Martin, and N. A. Rayner (2013), EN4: Quality controlled ocean temperature and salinity profi les and monthly objective analyses with uncertainty estimates, J. Geophys. Res. Oceans, 118, 6704–6716, doi:10.1002/2013JC009067
• JCGM, 2008: BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP, and OIML. Evaluation of measurement data — Guide to the expression of uncertainty in measurement. Joint Committee for Guides in Metrology, JCGM 100:2008. https://www.bipm.org/documents/20126/2071204/JCGM_100_2008_E.pdf/
• Lindstrom, E. , Gunn, J., Fischer, A., McCurdy, A. and Glover, L. K., A Framework for Ocean Observing. By the Task Team for an Integrated Framework for Sustained Ocean Observing, UNESCO 2012 (revised in 2017), IOC/INF-1284 rev.2, doi: 10.5270/OceanObs09-FOO
• Mieruch S, Demirel S, Simoncelli S, Schlitzer R and Seitz S (2021) SalaciaML: A Deep Learning Approach for Supporting Ocean Data Quality Control. Front. Mar. Sci. 8:611742. https://doi.org/10.3389/fmars.2021.611742
• Mieruch S, Kreps G, Chouai M, Reimers F, Vredenborg M, Rabe B, Tippenhauer S and Behrendt A (2025) SalaciaML-2-Arctic — a deep learning quality control algorithm for Arctic Ocean temperature and salinity data. Front. Mar. Sci. 12:1661208. https://doi.org/10.3389/fmars.2025.1661208
• Waldmann C, Fischer P, Seitz S, Köllner M, Fischer J-G, Bergenthal M, Brix H, Weinreben S and Huber R (2022) A methodology to uncertainty quantifi cation of essential ocean variables. Front. Mar. Sci. 9:1002153. https://doi.org/10.3389/fmars.2022.1002153
• Zhang, Qi, Chenyan Qian, and Changming Dong. “A Machine Learning Approach to Quality-Control Argo Temperature Data (2023).” Atmospheric and Oceanic Science Letters, Special Issue: Machine Learning Applications for Atmospheric and Oceanic Sciences, 16, no. 4: 100292. https://doi.org/10.1016/j.aosl.2022.100292.