PhD project in AI data analytics for future water security
Edinburgh Research Partnership in Engineering (ERPE) Joint PhD Studentship between Heriot-Watt University and University of Edinburgh.
Starting: September 2025
Application deadline: 18th February 2025
We are looking for a motivated and curious PhD candidate to work across Heriot-Watt University and the University of Edinburgh, UK.
Motivation
Climate change is influencing UK water resources, changing temporal and spatial patterns of water availability differentially across the country. Water resources is a data rich field, with long term datasets of meteorological and hydrological data available at different resolutions across the country. Recent research has explored the long-term influence of climate change on British rivers (Wray et al., 2024). The identified trends lead to exposure of regional response patterns. AI pattern recognition applied to spatial-temporal river runoff datasets will help to identify regional trends and the associated impact from climate change. Once trained, these algorithms can be deployed on future climate change projections (e.g. UKCP18 ensembles) to investigate how such patterns (e.g. rapid transitions between wet and dry) evolve in the future.
Aims and objectives
The overarching aim of this thesis is to explore future water resource patterns by examining current trends in hydrological extremes, and identifying regional occurrence of patterns; and using these understand future changes.
This PhD project will focus on application of modern AI algorithms to explore spatial-temporal data from UK rivers. It will entail profiling regional and temporal river runoff to explore spatio-temporal trends and identify extreme event occurrences. Feature selection will enable exploration of the driving stressors of hydrological events.
This is an exciting opportunity for the right candidate to tackle the challenge of climate change in water resources for future water security by analysing big data (national scale long-term hydrological data) using cutting edge data mining and AI tech to develop bespoke understanding and solutions.
Requirements
Engineering/physics/data science degree; ability to handle large datasets, coding skills in an advanced language (e.g., Python, or R).
Research environment and supervision
The successful candidate will be co-supervised by Prof Lindsay Beevers at the University of Edinburgh, Associate Professor Sandhya Patidar and Prof Vasily Demyanov at Heriot-Watt University as part of the Edinburgh Research Partnership in Engineering (ERPE). ERPE is a strategic alliance between Heriot-Watt University and the University of Edinburgh as two of the UK’s leading research universities in STEM. ERPE works with academics, industry and public sector partners to deliver world-leading engineering solutions and create commercial, social, environmental and economic impact.
How to apply
Contact us at l.beevers@ed.ac.uk, s.patidar@hw.ac.uk and V.Demyanov@hw.ac.uk by 18 Feb 2025. Please send your CV, transcripts and a motivation letter. The selected candidate will be invited to apply for the PhD position via the University of Edinburgh website by 16:00 on Friday, 28 February (UK time) and will be put forward to a competitive selection process by ERPE.
Comparison of the CC attribution percentage through time across eight first stage metrics for the four Tweed sub-catchments with longest records (from Wray et al., 2024).
Uncertainty quantification of geomechanically sensitive reservoirs using physics infused machine learning – CLOSED
A PhD opening in Uncertainty quantification of geomechanically sensitive reservoirs using physics infused machine learning. The position is fully funded from an industry research project (home tuition fee and stipend) for eligible UK candidates. International candidates will need to self-fund the additional tuition fee of around £20,000 per annum.
– CLOSED