The development of the system is articulated in many sequential tasks that allow to organize the work in specific working activities. In the beginning of the project, the user needs and requirements were examined, defined and analyzed. Feedback from the users was also gathered by composing an online questionnaire. Afterwards, the ontology scheme and the remote sensing algorithms that would be used were defined after a thorough investigation of the state of the art and in correspondence with the user requirements. The selected phenomena include hot spots, upwelling, phytoplankton blooms, fronts, oil spills, trophic status index, turbidity, winds, waves and shallow water places. Data mining algorithms for the EO image metadata were defined, followed by the establishment of the architecture design, connectivity and interoperability of the SEO-DWARF and the front-end of the web service design, taking into account the INSPIRE guidelines.
At a second stage of the work, the implementation of the service has been started. Currently, two main tasks have been achieved: 1. the remote sensing and data mining algorithms, and 2. the ontology with metadata formulation. It is stated that the aforementioned work was implemented during the secondments of the participants of the consortium.
As for the main results, two remote sensing pipelines were constructed: the first pipeline includes reading Sentinel 2 images, creation and application of masks for land and cloud removal, atmospheric correction, processing for producing a phenomenon product and modification of the INSPIRE file which contains the metadata. The second pipeline takes as input the SLSTR L2 product of Sentinel 3, extracts the bands of interests, processes them and stores the results into a GeoTiff image. The oils spills, upwelling and fronts phenomena have been implemented and a potential processing tool has been documented.
