England's water system faces a complex combination of ecological, infrastructural and social pressures. In January 2026, the government published its Water White Paper, describing it as "the biggest overhaul to water in a generation”, demonstrating the scale of the challenge at hand. In tandem, £104 billion of investment has been committed to the sector between 2025 and 2030, the largest in the industry’s history.
But legislation and investment alone cannot solve one of the deeper problems facing the sector. That problem is one of data, information, and knowledge.
There are significant amounts of data collected about the status of England’s rivers, lakes and seas by a variety of different actors, such as water companies, environmental groups and the government. However, this data collection is fragmented, and does not capture the full picture of England's rivers and waterways.
Our understanding of the long-term status of our waterways is built from data that is, structurally, incomplete. Because the current monitoring infrastructure was designed around fixed, periodic sampling, we cannot capture the dynamic nature of river health. While a pressure event might last for hours, the corresponding ecological shift might play out across months. It is in this context, and in response to these challenges, that citizen science groups have emerged, coming together to collect data to fill this gap.
A growing civic infrastructure
The growth of wild swimming, paddlesports and angling, combined with greater public awareness of water quality issues, has produced a generation of citizens actively engaged in monitoring their local waterways.
There are events like the Great UK WaterBlitz, coordinated by Earthwatch Europe, that mobilised 7,978 citizen scientists across 4,017 freshwater sites in a single week in April 2025. Surfers Against Sewage's citizen science programme now provides regular water quality testing at many inland sites. The Riverfly Partnership's Riverfly Monitoring Initiative trains volunteers to assess macroinvertebrate communities, with results capable of triggering formal Environment Agency investigations.
The civic infrastructure for community water monitoring is already here. And yet, despite its scale, citizen science data remains largely outside the formal mechanisms of water governance. It does not routinely inform regulatory decisions, and is rarely integrated with the official datasets held by the Environment Agency or water companies. There is significant value to this data, and we need to find a way to unlock its value.
Supporting the citizen science water data ecosystem
The ODI describes the different ways of empowering people to play an active role in the data ecosystem as participatory data. We see opportunities for more participation at different levels of the data ecosystem - at the data level, the organisational level and at the policy level. At the data level, people are generating and collecting data points about causes that matter to them. Citizen science is a key example of this, and initiatives like Zooniverse enable people to take part in real, cutting-edge research in many fields across the sciences, humanities, and more. Equally, initiatives like Ushahidi empower people through citizen-generated data to develop solutions that strengthen their communities, for example supporting people through natural disasters. Programmes such as Impetus provide resources and funding for citizens to set up new initiatives to collect data and use it to drive meaningful change. Initiatives such as the EU Prize for Citizen Science or the UKRI's Citizen Science Exploration Grants encourage and incentivize public participation in research and policy making.
At the same time, the ODI has been working with Stream, a collaboration between water companies in the UK, supported by industry and civil society partners. Members of Stream have a vision to unlock the potential of water data to benefit customers, society, and the environment, and to use data to address key water challenges. Stream has the ambition to become an independent data institution, an organisation that stewards data on behalf of others, often towards societal, economic and environmental benefit. Initially Stream focused on publishing data openly, but moving forward there is an emerging role to play in empowering people to have more control over data.
Over the past 12 months, we’ve been working together to understand how we can best support the citizen science water data ecosystem, to realise the significant value that citizen science data holds for the water data ecosystem. Through a series of three blog posts we detailed our research on the challenges and opportunities facing citizen science data in the UK. We then hosted a design sprint at the Northumbrian Innovation Festival, with experts from across the sector, which produced the first iteration of the trust data sharing framework, and two draft data ecosystem maps, demonstrating the necessity of collaboration in tackling data sharing challenges in any sector.
We aim to bring our collective experience and expertise from working with Stream and partners in respect of open data publication (governance, standardisation, use cases etc) to help realise the full potential of citizen science data, and will be working closely with other organisations in this space.
Building trust across the water sector
We know that there is a breakdown in trust between citizen science groups, and those who they seek to influence with data. This trust deficit is compounded by limited funding for citizen science programmes, a narrow understanding of the potential value of community-driven environmental monitoring, and a lack of visibility of the breadth of data collected by citizen science groups which can support a more holistic view of a river's health.
However there are exciting new possibilities emerging. We know citizen science data is, by its nature, spatially dispersed, temporally variable and methodologically mixed, which has made it difficult to integrate with institutional monitoring. But machine learning models trained on validated datasets can now contextualise volunteer observations in ways that were previously impractical. Projects like River Deep Mountain AI, funded through Ofwat's Innovation Fund, are developing open-source machine learning models that integrate sensor data, satellite imagery and citizen science to track river health trends across pollution sources. Led by Northumbrian Water, with partners including The Rivers Trust and Google, it is a strong example of what AI-assisted integration could look like at scale. A peer-reviewed study on the River Chess showed that machine learning models trained on citizen science data could explain water quality dynamics, including the influence of nearby wastewater treatment works, with less than 1% error. While Ofwat are actively considering how citizen science data can help provide additional evidence to demonstrate environmental outcomes delivered water industry investment, through project Clear Waters.
AI is not here as a replacement for community monitoring. It is the layer that supports the utility and legibility of community data, the missing connection between observation and action. Ultimately, there is a need to make citizen science data AI-ready, as a step towards realising its full potential.
In the next 9 months, we will be working to further develop our first iteration of the trust framework. We believe the co-creation of the framework will build bridges between the different stakeholders in this ecosystem, and ultimately provide one of the stepping stones towards realising the value of citizen science data. This will involve:
1) A better understanding of the different actors in the water sector.
The water data ecosystem is complex, and has many factors impacting the quality of the water, and wildlife. While we have initially mapped out two data ecosystems, there are still other actors often left out of these conversations. To build trust across the ecosystem, we need to first understand who the different actors are, such that we can bring them into the conversation.
2) User research with the stakeholders identified in the ecosystem map to understand what their needs are from citizen science data.
We need to better understand how this data can support their aims, and where it may provide additional value, such that we create a trust framework which works for everyone.
3) Co-designing the trust framework
Bringing together actors from across the ecosystem at different events to work together to co-design and further develop the trust framework, to build trust, better understand each parties hopes and concerns, and to ensure the framework meets everyone's needs.
4) Exploring the different funding models to support citizen science data to thrive
Citizen science is not free, for example there are costs associated with purchasing kit, testing samples, hosting data and much more. We will research what a funding model for supporting this ecosystem looks like, building on our work on sustainable data access models.
5) We’ll be updating and adapting our current v1 of the trust framework
Building on the insights learned through this research, and through testing the framework on two use cases:
- eutrophication (the Eutrophication Assessment Weight of Evidence (WoE) Tool is a decision support tool used by the Environment Agency to assess where rivers are suffering from eutrophication and to guide where further evidence or action is needed)
- e-coli (a bacteria found naturally in the environment, however certain strains or excessive amounts can lead to illness).
6) Finally, we will publish an updated version of the citizen science water data trust framework
If you’re interested in learning more about this project, or would like to get involved with this work, get in touch with us at [email protected].
Photo by Nick Fewings on Unsplash