Democratizing Hydrological Data:
Our Commitment to Open Science
Solving the global water crisis requires more than just innovative technology, it requires transparency, collaboration, and shared knowledge. As an EU-funded Horizon Europe initiative, the SWIM project is fundamentally committed to Open Science. We believe that providing open access to our research, machine learning models, and datasets empowers a global community of water authorities, researchers, and citizens to make proactive, data-driven decisions.
The Standard: All our shared outputs adhere strictly to the FAIR principles, ensuring our data is Findable, Accessible, Interoperable, and Reusable.
Explore Our Public Deliverables
Track our progress and explore the foundational strategies driving the SWIM platform. The following official project deliverables are available for public viewing and download:
-
D2.1 covers the design and implementation of SWIM’s EO and in-situ data ingestion pipeline, explaining how satellite Earth Observation data, WAMO sensor measurements, and third-party environmental datasets are acquired, pre-processed, harmonized, validated, and stored for use across the SWIM platform. It details the data sources, including Copernicus, CDSE, LP-DAAC, Earth Explorer, WAMO, and local datasets, and explains how these inputs support the three core SWIM modules: water quality monitoring, water balance assessment, and natural disaster assessment. The deliverable also describes the role of OPIE and the WOLF calibration/flagging system in improving data reliability, traceability, and confidence, while outlining the technical infrastructure, cloud and edge computing approach, and machine-learning readiness of the pipeline.
-
D3.1 covers the Comprehensive Machine Learning Solution Package developed for SWIM’s Water Balance module and wider AI-DSS framework. It explains how the project uses machine learning to estimate and predict streamflow in poorly gauged or ungauged river basins by combining global Earth Observation and hydrological datasets, including ERA5-Land, GloFAS v4.0, and HydroSHEDS. The deliverable documents the full ML workflow, from automated watershed delineation and data preprocessing to feature engineering, model training, validation, and comparison of Support Vector Regression, Random Forest, and XGBoost models. It also describes the reusable SWIM.py module, Jupyter notebook tools, performance metrics such as NSE, R², PBIAS, and RMSE, and how the outputs support future flood, drought, water availability, reporting, and decision-support functions within the SWIM platform.
-
D5.2 covers SWIM’s Dissemination, Exploitation and Communication Plan, setting out how the project will build visibility, share results, engage stakeholders, and prepare the platform for future market uptake. It explains the project’s communication channels, including the SWIM website, LinkedIn, social media, webinars, newsletters, videos, promotional materials, and partner networks, while also defining the visual identity and brand presence used across public-facing materials. The deliverable also outlines dissemination activities such as publications, events, open data sharing, and stakeholder engagement, alongside exploitation planning through the future Business Innovation Plan, Commercialisation Plan, and roadmap from TRL4 toward higher technology readiness. Overall, it positions SWIM as a credible, scalable water management solution by connecting technical progress with public awareness, end-user adoption, policy relevance, and future commercialisation.
-
D5.3 covers the creation of SWIM’s one-minute concept demo video, designed as a strategic communication asset to explain the project’s vision, value, and core capabilities in a clear, accessible format. It documents the video’s purpose, target audience, production approach, audiovisual content, and dissemination role within WP5, including how the video combines voiceover, motion graphics, stock footage, AI-generated visuals, and SWIM interface screen recordings to show the integration of Copernicus Earth Observation, WAMO IoT sensors, EGNSS positioning, AI analysis, alerts, dashboards, and reports. The deliverable also outlines how the video supports stakeholder engagement, public visibility, investor and regulator interest, and broader outreach through the website, social media, events, and presentations.