The Global Fisheries Atlas Virtual Lab (GFA VLab) will provide FAIR-compliant services for end users to better discover, access andunderstand available data and knowledge about fisheries worldwide.GFA VLab will combine Global Record of Stocks and Fisheries (GRSF) andthe Fisheries Atlas (FA) VLabs (developed in the previous Blue-Cloud Pilotproject), thereby integrating and harmonising new datasets relevant forfisheries variables. These VLabs will be an opportunity to combinefisheries data and environmental parameters at a global scale. Theunderlying R-based data workflow will be strengthened to become morerobust and more generic. DOI assignment as well as extensive use of OGCstandards for spatial (meta)data as well as CWP standards for fisheriesdata management is made to foster data discovery, interoperability, andreuse. Indeed, these standardized data formats and access protocolsfacilitate the reuse of the datasets while DOI ensure a betterdissemination, long-term availability and citation. Semanticinteroperability is also facilitated by implementing best practices forknowledge management (semantic Web standards).

The GFA VLab is helping to provide FAIR gridded data at a global scalewhich can be used to explore the benefits of AI for fisheriesmanagement. In particular, the GFA data can be used to train AImodels which predict tuna species distributions by learning, fromhistorical data, what are the best environmental conditions for thesespecies. For example, by combining fisheries data time series of the TunaAtlas (providing gridded maps of catches and fishing efforts) or speciesoccurrences (from fishing operations) with environmental data (from e.g.Copernicus remote sensing or models outputs data). This is expected tohelp researchers uncover new and unexpected correlations between theoceanic conditions and species in the open ocean.

This poster was showcase at the 2024 Fisheries Science Seminar – Artificial Intelligence. More info: https://blue-cloud.org/events/2024-fisheries-science-seminar-artificial-intelligence