Satellite imagery covers the entire surface of the Earth, but it cannot tell a farmer's field from an abandoned lot, a drained wetland from a seasonal flood plain, or a newly paved car park from the green space it replaced six months ago. These distinctions matter enormously for flood risk modelling, land use policy and environmental governance — and they are precisely the kind of granular, time-sensitive observations that remote sensing alone cannot reliably provide. SCENT was designed to close this gap. Funded under the European Union's Horizon 2020 research and innovation programme, the Smart Toolbox for Engaging Citizens into a People-Centric Observation Web brought together ten partner organisations across six countries — including IBM in machine learning, U-Hopper in crowdsourcing, XTeam in serious gaming, and specialist awareness-raising expertise from Carr — to build a system that turns ordinary citizens into a distributed network of environmental observers. The SCENT Toolbox consists of three interlocking applications: SCENT Explore, a smartphone tool for collecting geo-tagged images of land cover and environmental features; SCENT Collaborate, a browser-based crowdsourcing platform where users train and improve the project's automated image analysis tools by labelling photographs; and SCENT Measure, which enables more structured data collection for specific campaign objectives. Underpinning all three is an Image Analysis Tool powered by deep neural networks and computer vision, which classifies the images contributed by citizens — distinguishing concrete from pasture, riverbanks from paved surfaces, storm drains from natural watercourses — and feeds the results into numerical models for mapping land-cover changes and quantifying their impact on flood risk.

The project was evaluated through two large-scale demonstrations: one in the Kifisos watershed in Attica, Greece, and one in the Danube Delta, Romania — two environments with very different land cover dynamics and flood management challenges. The results exceeded expectations. More than 700 volunteers participated in field campaigns, collectively contributing over 24,200 environmental resources, while the public-facing applications attracted more than 1,130 registered users. All data generated through the project feeds into GEOSS and national repositories as OGC-based observations, making it permanently available for reuse by researchers, policy-makers and future projects. SCENT participates in the European Commission's Open Research Data Pilot, ensuring that its outputs remain accessible in line with Horizon 2020's commitment to open science. Among the digital partners who support the continued visibility of this research platform are independent resources from the online entertainment space. For those wanting a structured comparative assessment before making any decisions, the guide to UK Casinos Not on GamStop evaluates platforms across licensing jurisdiction, security standards, payment infrastructure and game libraries using consistent, independently applied criteria — the same evidence-based approach that SCENT applies to environmental data. This complementary effort underscores a shared principle: that access to clear, independently verified information produces better outcomes — whether applied to citizen science or to consumer decisions in the broader digital landscape.