Inspiration
Urban streams connect environmental health with everyday community life. Heatwaves, polluted runoff, and degraded riverbanks can affect aquatic biodiversity while also increasing risks for people. Inspired by the One Health approach and the IEEE OneAquaHealth hackathon’s Data-to-Insight track, we wanted to make these connections easier to understand and act on.
What it does
OneAqua Insight Hub is an interactive decision-support dashboard covering five European pilot cities: Coimbra, Toulouse, Benevento, Gent, and Oslo.
It brings ecological health, human health risk, and urban climate stress into a shared view, summarized through an overall One Health Vitality Score. Users can explore stream maps, environmental trends, correlation charts, and early-warning indicators for oxygen depletion, microbial spikes, and conditions associated with mosquito breeding.
A scenario sandbox lets users explore heatwaves, storm runoff, and riverbank vegetation restoration. Community members can submit observations, which receive a basic rule-based confidence assessment. The dashboard also generates downloadable policy briefs with suggested interventions.
How we built it
We built the application in Python using Streamlit, pandas, and NumPy. Folium powers the interactive maps, while Plotly handles time-series charts and correlation heatmaps.
The data layer incorporates downloaded records from Hub’Eau, GBIF, and ARPAC alongside modeled baselines for Coimbra and Oslo. Demonstration sensor locations and modeled indicators fill gaps in the prototype. A weighted scoring engine combines ecological, public-health, and climate indicators, while separate components handle mapping, analytics, reporting, and scenario controls.
We deployed the application on Streamlit Community Cloud so anyone can explore it without installing anything.
Challenges we ran into
The biggest challenge was bringing different kinds of environmental data into a consistent format. Water measurements, biodiversity observations, and ecological assessments differ in structure, coverage, and sampling frequency.
We also had to balance a useful demonstration with incomplete data, normalize indicators with different scales, and keep the dashboard responsive. During development, we improved data caching, corrected health-risk calculations, resolved scenario preset synchronization, and worked through map tile configuration issues.
Accomplishments that we're proud of
We’re proud of delivering a deployed prototype that connects environmental exploration with practical decision support. Users can move from a mapped stream segment to broader trends, explore a possible intervention, and export a readable advisory.
The scenario sandbox is a particular highlight: it makes the relationships between climate pressure, vegetation, water quality, and community health tangible.
What we learned
We learned that turning data into insight requires more than charts. Data provenance, consistent scales, clear explanations, and honest treatment of missing information all matter.
We also learned how much interface design affects understanding. A clear lead metric, focused views, and plain-language recommendations help users navigate a complex environmental problem.
What's next for OneAqua Insight Hub
Our next steps are to connect live monitoring feeds, expand measured-data coverage, and clearly distinguish observed values from modeled estimates. We also want to validate the scoring model with domain experts, match citizen reports to nearby monitoring stations, and add persistent report storage and alert history.
Ultimately, we want to develop the prototype into a useful tool for local authorities, researchers, and communities caring for urban freshwater ecosystems.
Log in or sign up for Devpost to join the conversation.