PhD: Machine learning enhanced sea-ice dynamics for climate modelling via FindAPhD

University of Reading

Reading, UK 🇬🇧

About the Project

This PhD is part of the Net Zero Polar Science DTP, which aims to make polar science possible in a net zero world. For further details visit https://nzps-dtp.ac.uk/

Sea ice, formed from the freezing of ocean water, consists of wide (10-5000m), thin (0.1-5m) plates called floes that freeze together in winter and melt apart in summer. Sea ice plays a central role in the climate system, limiting air-ocean transfers of heat, moisture and momentum, and moderating global atmospheric and oceanic circulations and extreme weather. The unprecedentedly rapid decline of Arctic sea ice is a conspicuous indicator of climate change, and there are recent dramatic losses in the Southern Ocean. The state of the sea ice cover is important for navigation, wildlife, and indigenous communities.

Climate sea ice models simulate the melting, freezing, movement, and deformation of the ice cover. Wind and ocean stresses cause deformation, they can (i) compress the floes, causing them to break apart and pile up to form pressure ridges and keels, (ii) cause the floes to slide along their edges, or (iii) separate the floes to form open water areas [Feltham, 2008]. The sea ice rheology module calculates the internal ice forces that resist deformation and is a computational bottleneck in numerical simulations of sea ice due to the need to iterate stresses to convergence.

The chief innovation in this project is to replace an existing rheology module with a physically informed Machine Learning (ML) algorithm that is computationally efficient and therefore consumes less carbon to run. Such an algorithm would reduce the computational cost of the sea ice model, allowing for more accurate sea ice simulations, and offer a route to include more sophisticated rheology theory. Preliminary results [Thomas and colleagues, unpublished] have shown that a ML algorithm demonstrated skill in predicting sea ice velocity. The ML algorithm will be incorporated into a climate sea ice model and its impact assessed as part of a Net Zero case study. Since sea ice models are used operationally and for climate projections, e.g. for the Intergovernmental Panel on Climate Change, the ML algorithm could produce significant energy and carbon savings.

Our specific objectives are:

1. Run a state-of-the-art climate sea ice model to build a novel training dataset of sea ice stresses

2. Develop a machine learning based emulator of sea ice dynamics

3. Test the performance and carbon savings of the dynamics emulator in climate model simulations

Net Zero Case Study: The student will partner with the Met Office to work on sea ice modelling. The machine learned model of sea ice rheology will be brought into a representative sea ice modelling configuration and used to simulate past and future sea ice changes. Since the sea ice rheology is the most computationally costly component of the sea ice model, and this is used operationally and for climate projections, e.g. for the Intergovernmental Panel on Climate Change, this has the potential for significant energy and carbon savings, which will be assessed.

Technical Approaches

Existing models of sea ice rheology are computationally inefficient due to the need to iterate to convergent stresses. Our chief innovation is to replace the rheology module with a machine learned algorithm that is computationally efficient and therefore consumes less carbon to run.An existing rheology module will be ran offline at high resolution to convergence, recording output stresses for all conceivable inputs, to generate a comprehensive dataset used to train the ML algorithm. The ML algorithm would be embedded into the sea ice model and its impact on simulations and carbon cost evaluated.

Training Opportunities

The successful applicant will join a team of scientists working on a range of allied projects on Arctic and Antarctic sea ice, who will provide support on technical aspects of model development, simulation and analysis. The student will acquire skills in computational techniques. In addition to substantial training opportunities in the Department of Meteorology at the University of Reading, the student will spend time at the British Antarctic Survey and Met Office and gain experience of working in a large research organisation with a strong focus on delivery of operational products and services.

Academic Background

The successful candidate will have a degree in physics, applied mathematics, computer science, engineering, or a similar numerate subject such as meteorology, along with an aptitude and enthusiasm for solving real world problems and computer programming. The candidate should be able to work independently and in groups and be enthusiastic. While knowledge of the climate system, oceanography, sea ice, or meteorology would be ideal, this knowledge can be taught to the right candidate.

Eligibility

For entry to PhD study, applicants are expected to have at least one of the following:

• a first or upper second (2:1) class honours undergraduate degree in a relevant subject, or an equivalent international qualification,

• a relevant master’s qualification or equivalent evidence of prior professional practice.

International applicants and candidates from non-English speaking countries will need to meet the minimum language requirements for admission onto the programme of study for their Home institution.

How to Apply

Informal enquiries about the project and your application should be addressed to the project supervisor, Professor Danny Feltham, d.l.feltham@reading.ac.uk

After you have discussed your application with the project supervisor and read the NZPS application guidance, you should:

1) Complete the online NZPS Application Form by 10.00 (GMT) Wednesday 6th January 2027

2) Submit any additional application documents in the requested format to nzps@northumbria.ac.uk by the closing date.

If you require any additional assistance in submitting your application or have any queries about the application process, please don’t hesitate to contact us at nzps@northumbria.ac.uk


Funding Notes

Available to Home and EU/International students, subject to successful completion of quality assurance checks and UK Visa and Immigration (UKVI) compliance requirements. This includes full stipend at UKRI rates (£21,805 p/a for 2026/27 full-time) and Research Training and Support Grant.

Full tuition fees covered for Home/UK students and a contribution towards fees, up to Home/UK amount, is provided for international students. International students may apply for this project, but are required to meet the difference between international and home fees from an alternative funding source. Please note: additional costs may apply for international applicants, including visa costs and NHS surcharge.


References

Feltham, D. L. (2008) Sea ice rheology. Annual Review of Fluid Mechanics, 40 (1). pp. 91-112. ISSN 1545-4479 doi: 10.1146/annurev.fluid.40.111406.102151

44 days remaining

Apply by 6 January, 2027

POSITION TYPE

ORGANIZATION TYPE

EXPERIENCE-LEVEL

DEGREE REQUIRED

Register for World Toilet Summit 2026 - Thailand - November 17-19, 2026