By Rosie Perrett, Dr Connie Tulloch, and Matt Coombs
The Healthy Waters blog is back! Our research group are currently working on three government funded research projects monitoring the health of rivers using sensors to measure various biological and physiochemical parameters.
In 2022, no rivers in the UK were classified as having good overall health, which can be largely attributed to discharges of sewage effluent and agricultural runoff into rivers (The Rivers Trust, 2024). Collecting water quality data allows us to detect and monitor the impacts of pollution events in rivers and can inform evidence driven decision-making to improve river health and water safety.
All three projects are using in-situ sensing technologies and spot sampling (paired with laboratory analysis) to assess UK river quality. However, each project is measuring slightly different water quality parameters according to the project aims and deliverables.

Collecting water quality data with Chelsea fluorescence-based sensors and Xylem sondes
Development of an Innovative Intelligent Multiparameter Fluorometer to Sense the Impact of Organic Pollution on River Health
Project partners: University of the West of England (UK academics), Chelsea Technologies (UK sensor manufacturers), The Rivers Trust (river conservation experts)
This project is funded by Innovate UK and aims to develop and deploy a new multiparameter fluorescence-based sensor measuring organic pollution, algae (phytoplankton), and bacterial contamination. The new sensor technology will provide data on river health, for which there is currently a lack of data on since commercially available sensors predominantly measure physiochemical parameters.
Following sensor development, prototypes will be deployed at three locations in the River Dart, collecting water quality and river health data in real time. River water samples will be collected for laboratory analysis involving: measurements of BOD5, microbial enumeration (heterotrophic bacteria and fecal indicator organisms), as well as carbon and nutrient concentrations.

Another field trip with the Chelsea and Xylem sensors
The sensor will be low cost, making it accessible for community groups. AI and machine learning will be used to create a Water Quality Index by combining the measurement of the new parameters with additional real-time data, enabling interpretation of sensor data regarding river health.
This new sensing technology will enable accurate, high-resolution monitoring of the impact of pollution discharges on bacterial contamination and organic pollution in rivers, improving the way we monitor and determine river health.
MaD-OPS project – Monitoring and Detection of Organic Pollution from Sewage in rivers.
Funded by the Natural Environment Research Council (NERC), the MaD-OPS project aims to investigate the impact of pollution from both point sources (like sewage discharges) and diffuse sources (like agricultural run-off) on river health.
Using fluorescence-based sensors supplied by our industry partner Chelsea Technologies and multiparameter Xylem EXO sondes, we will monitor a range of physical, chemical and biological water quality parameters.
Our sensors will be connected to custom WATR tech environmental monitoring platforms, and data will be continuously uploaded and accessed through a cloud-based system.
This forms the basis of our sensor network to monitor river health in our demonstrator catchment – The Bidwell Brook.

Bidwell Brook, Devon
Alongside continuous sensor data provided to us via our sensor network; we’ll also be taking regular spot samples for ground truthing. Looking into nutrient concentrations, total carbon, biochemical oxygen demand (BOD5), faecal indicators, as well as DNA extraction for microbial communities.
By linking our data with hydrological data – provided by the Environment Agency and event duration monitoring data (EDM) – provided by South West Water, we will create a Water Quality Index using AI and machine learning. This will provide a platform for communities to assess the health of their local rivers and see the impact of pollution events in real time, empowering them to take action when necessary.
TWIN–Waters: Management of water resources: resilience, adaptation and mitigation to hydroclimatic extreme events and management tools.
This project is funded by the European Commision and UK Research and Innovation (UKRI) with support and collaboration from the University of Lund (Sweden), The University of The West of England (UK) and Wrocław University of Environmental and Life Sciences (Poland).
The Project aims to integrate AI and Machine Learning with GIS and large data sets to create a digital platform for water quality monitoring. This digital platform will act as the “Digital Twin” of the water source which will be a valuable management tool.
There are three phases to the project with UWE primarily involved in providing data from a range of novel sensors being used across other similar projects (MaD-OPS & Multiparameter Fluorometer).
The project is currently in Phase One which aims to identify monitoring sites in each of the three countries. At these sites, water quality sensors will be deployed to collect data which will be used in the GIS modelling alongside satellite data.
In addition to this, each university partner will print and build 8 3D-Printable open-source microscopes designed by OpenFlexture. These microscopes will be used identification of Algae species, citizen science and public engagement events.

3D microscopes at the Festival of Nature
UWE have seven monitoring sites across the River Dart and Bidwell Brook, from which data from sensors and traditional water quality parameters from laboratory analysis are being collected. The same sample sites are being used as the other projects within our research group as this will allow cross project data sharing, ultimately leading to a more comprehensive understanding of the catchment.
The next stage of the project is to develop a functional model for each of the three catchments using the sensor data coupled with satellite data. This will require regularly uploading data to the digital platform for the AI and Machine Learning aspect of the model. The model will then begin to make assumptions and eventually make predictions to allow accurate water quality management.
We’ll be posting on the Healthy Waters blog every month, with our next post being an introduction to the team.














