Decision on the EOVs to be used as a pilot and identification of the available datasets (MS2)

This milestone is a report on the preparation process for the decision on the Essential Ocean Variables (EOVs) to be used as a pilot in the ObsSea4Clim project and on identifying the available datasets. An adequate amount of data is a prerequisite to train
any Artificial Intelligence/Machine Learning (AI/ML) model to perform well a desired task and the choice of a pilot EOV could depend primarily on that. However, by involving the consortium partners in this decision with an internal survey, we have
achieved the following:
● agree on a pilot EOV that is also fit-for-purpose for a number of ObsSea4Clim partners’ needs,
● test on an additional second variable and ensure consistency in EOVs of interest throughout the project,
● focus on testing the AI/ML algorithms in a smaller region, since the global scale approach will pose serious computational limitations.
Furthermore, since the performance of the AI/ML models depends on the data used to train them, the data availability of the EOV of interest is also considered together with the partners’ feedback. This report briefly explains the rationale behind the choice of the pilot EOV based on the literature and more importantly, on the partners’ contribution.