Data science

The SEFAP data pipeline will comply with FAIR data principles and will build on existing standard operating procedures in AQUACOSM and AQUACOSM-plus (Knowledge Base | AQUACOSM).
FAIR data: Overseen by DANS-KNAW, our data management plan [SM1] outlines how we will secure primary data, harmonize data curation through Standard Operating procedures (SOPs), and publish mesocosm data. Training and documentation shall be provided on state of the art research data management procedures.[2] To encourage Findability of datasets, SEFAP will transfer and maintain the existing metadata catalogue of AQUACOSM-plus.
Data resurrection: In addition to new experimental data, SEFAP will FAIRify unpublished data from previous experiments.
Data vizualization: SEFAP will improve the existing near-real-time data visualization platform of the Limnotrons to harvest additional automated datastreams from partners.
Data processing: Through University of Twente, SEFAP will implement advanced machine learning algorithms for innovative signal processing to ensure reliable, robust, and tractable data streams. To process the wide array of data types UT's upgrade will allow the fusion of multimodal (sensor) data, accommodating variations in resolution, type, and dimensionality.