PhD Studentship: Strategic Optimization of Machine Learning-based Metocean Forecasts, NERC GW4+ DTP PhD studentship for September 2025 Entry Ref: 5420 Department ESE

About the Partnership

This project is one of a number that are in competition for funding from the NERC Great Western Four+ Doctoral Training Partnership (GW4+ DTP).  The GW4+ DTP consists of the Great Western Four alliance of the University of Bath, University of Bristol, Cardiff University and the University of Exeter plus five Research Organisation partners:  British Antarctic Survey, British Geological Survey, Centre for Ecology and Hydrology,  the Natural History Museum and Plymouth Marine Laboratory.  The partnership aims to provide a broad training in earth and environmental sciences, designed to train tomorrow’s leaders in earth and environmental science. For further details about the programme please see http://nercgw4plus.ac.uk/

Project details

For information relating to the research project please contact the lead Supervisor via A.Pillai@exeter.ac.uk

Project Aims and Methods

This project partners a pioneering academic research team with the UK’s National Weather Service, to develop innovative methods for training machine learning (ML) weather prediction frameworks to improve meteorological and oceanographic (“metocean”) forecasts for marine and maritime applications. Traditional ML approaches learn from extensive datasets, which can be computationally demanding to use. This PhD research will therefore develop new strategies to optimize training, reducing data requirements while maintaining accuracy, enabling new approaches for regional metocean prediction. 
As data-driven weather prediction evolves, ML models trained on historical data could surpass traditional numerical weather prediction (NWP) methods in accuracy and efficiency. However, the assumption that longer training datasets improve model performance faces challenges, particularly in extreme conditions. The project will investigate optimizing training data quality over quantity, potentially reducing the computational cost and improving the pace of model development/deployment in application. 
The student will review existing ML practices in meteorological and oceanographic forecasting, benchmarking current methods, and proposing new training techniques. These may include varying the length of training data, using smaller models for specific seasons or regimes, and generating synthetic datasets. This research aims to disrupt current ML training practices by creating more efficient and accurate schemes for metocean forecasting. 

Project partners

The Met Office are meeting the extra expenses (such as travel and subsistence) incurred by the student visiting and working at the Met Office. They are also contributing to cash or in kind towards necessary materials whilst the student is based at the Met Office.

Training

The DTP offers funding to undertake specialist training relating to the student’s specialist area of research.

Advert information