PhD Studentship - Harnessing Mussel Behaviour and Machine Learning for Coastal Water Quality Monitoring

Location: Streatham Campus

Project details:

The health of marine ecosystems is increasingly threatened by pollution, eutrophication, and harmful algal blooms (HABs), which have profound ecological and socio-economic impacts. Bivalves, such as mussels, are highly sensitive to water quality changes and exhibit distinct valve gape behaviours in response to environmental stressors including oxygen depletion, toxic algae, and pollutants. This natural sensitivity makes them powerful bio-sensors for environmental monitoring, capable of providing early warnings of ecosystem stress. However, harnessing this behaviour in a practical, real-time monitoring system requires advances in both sensor engineering and behavioural data interpretation.

This PhD project aims to develop a next generation environmental biosensor, transforming an existing prototype technology into a deployable environmental monitoring solution. Building on recent innovation at Exeter, this PhD will translate a novel discrete gape-sensor unit, valuable for laboratory and short-term field studies, into a fully integrated system for real-world deployment, where rapid detection of water quality deterioration is critical. The proposed research will expand current technology to include automated live analysis, integrating machine learning algorithms capable of interpreting the complex behavioural patterns of mussels in response to environmental stress. This will enable generation of real-time alerts and warnings, allowing for rapid response to pollution events or the onset of HABs. The technical innovation lies in combining robust low-power hall-sensor hardware with wireless communication and complex analytical software. Mussel behaviour is not binary (open/closed) but highly dynamic, with subtle patterns indicating different stressors. Training machine learning models to distinguish between normal physiological behaviour (e.g. diurnal rhythms, feeding responses) and abnormal stress-induced patterns will be central to the project. This requires interdisciplinary expertise: biosciences to characterise the physiological responses of mussels under controlled exposures, and engineering to design hardware, firmware, and analytical pipelines that can run autonomously in the field. The societal and industrial value of this system is significant. Coastal communities and industries such as aquaculture are vulnerable to HABs and pollution events that can cause mass mortalities, economic loss, and human health risks through contaminated seafood. A low-cost, deployable sensor network based on mussels could provide real-time environmental intelligence, supporting regulatory agencies, aquaculture managers, and marine spatial planners. Furthermore, as climate change drives increases in the frequency and intensity of HABs, tools that enable proactive management will become increasingly vital. The project will be delivered via multidisciplinary collaboration, offering a unique training environment. Dr Robert Ellis (Biosciences) will provide expertise in bivalve physiology and aquaculture, ensuring robust experimental validation of behavioural biomarkers. Dr Jun Chew (Engineering) will lead on sensor design, system integration, and embedded software development. Industrial engagement is provided by Prof Mike Allen and SeaGen, a blue-tech company based in the South West of the UK, who will support product development and route-to-market strategies. Together, the supervisory team will provide a unique environment bridging fundamental biology, applied engineering, and industrial innovation. In summary, this PhD will deliver both fundamental insights into bivalve behavioural ecology and a technological platform with transformative potential for environmental monitoring.

Please direct project specific enquiries to: Contact primary supervisor (r.p.ellis@exeter.ac.uk) Please ensure you read the entry requirements for the potential programme you are applying for. To Apply for this project please click on the ‘Apply’ button above.

Advert information