Call for papers! Submit your paper to the "International Journal of Data Science and Analytics" before the 30th of June 2024!
This demonstrator is jointly developed by the Information and knowledge management Team (NFISI) of the Fisheries Division, Food and Agriculture Organization of the United Nations (FAO) and the French provider of environmental monitoring solutions Collecte Localisation Satellites (CLS).
The ambition is to deliver a tool to produce national aquaculture sector overviews whereby a country can make use of OGC compliant data services to monitor its aquaculture sector, not in an isolated way, but built on interoperable services where teams can compute and publish reproducible experiments.
- End-users: general public with an interest in aquaculture locations, production, and tracking, accessed through a public web-portal.
- Advanced users: regional aquaculture data analysts that need to monitor how aquaculture in their area of interest develops over time using Sentinel and other satellite data, and bring that in relation to environmental variables and other ancillary data such as site-inventories.
- Developers: system developers in need of a ‘template’ solution for the management of sentinel and other satellites data access and processing in WeKEO. The demonstrator example workflow on cage monitoring can be adapted to other data and analytical WeKEO processes.
- Regional Aquaculture Managers: regional aquaculture managers who require access to overviews of aquaculture sites and area estimates to inform their management decision making processes.
An aquaculture monitor can become an important information source for local and national governments that lack the capacity to implement a national monitoring tool. It can thus provide a monitor for several Sustainable Development Goals such as 14 "Life below water", and 2 "Food Security".
The Jupyter notebook for aquaculture cage detection generates GEOPACKAGES over an ROI using Copernicus Sentinel 1 images. The first step is a tiling service to prepare the data for analysis, while in the second step the cages are detected. The output is ingested the Spatial data infrastructure that supports the VLab and is managed through D4Science, and shown in the ISO/OGC compliant Map viewer in the Aquaculture VLab.
The VLab ingests AI based land-types classifications as GEOPACKAGES over an ROI that provided the base for a validation based on in-situ data. The Blue-Cloud approach showed the technical feasibility to interoperate with external proprietary software and bring the results in a collaborative environment. The results can be mashed up with other Blue-Cloud products.