Research Fellow in Machine Learning for African Weather Prediction

Location: Leeds - Main Campus

Working time: 100% - We are open to discussing flexible working arrangements

Contract type: Fixed term (until 30^th September 2027 to complete specific time limited work)

Interview date: To be confirmed

This role will be based on the university campus, with scope for it to be undertaken in a hybrid manner. We are also open to discussing flexible working arrangements.

Are you a Computer Scientist looking to apply your expertise to real-world weather forecasting challenges in Africa?

Machine-learning has the potential to revolutionise weather prediction for and in Africa, and we are seeking a machine-learning specialist ready to embrace challenges in weather prediction, with the aim of creating operational systems benefitting millions of people across Africa. 

The Cumulus project is a consortium of UK and African partners funded by the Gates Foundation, which aims to make a breakthrough in the application of machine-learning forecasting methods for West African agriculture. The project is led by the UK’s Alan Turing Institute, with partners in Senegal and Ghana, and all partners will collaborate closely. We will also be part of an over-arching project – Nimbus – linking with East African teams and other international specialists.

Within Cumulus, you will lead the development of “downscaling” methods for sub-seasonal (2-4 week) forecasts. Our priority will be to implement deep-learning based methods, to turn global sub-seasonal predictions (which are at quite coarse spatial resolution and have systematic biases) into accurate, actionable information for farmers at their “field scale”. The machine-learning methods will include training with data from Africa not normally available for routine forecasting, and we will explore the value of local optimisation or fine-tuning of foundation codes. Handling spatio-temporal statistics of forecast uncertainty will be a key consideration. This kind of downscaling with machine-learning methods is a rapidly advancing field and it is an exciting time to influence the development of the methods. We aim to get the first codes developed rapidly, and to ensure that code can be run, evaluated and improved by our partners in African universities and weather services in Senegal and Ghana. Evaluation and benchmarking of the predictions will be a major priority, working with colleagues across the project and across Africa.

Please note that this post may be suitable for sponsorship under the Skilled Worker visa route but first-time applicants might need to qualify for salary concessions. For more information please visit: www.gov.uk/skilled-worker-visa.

For research and academic posts, we will consider eligibility under the Global Talent visa. For more information please visit: www.gov.uk/global-talent.

What we offer in return

• 26 days holiday plus approx. 16 Bank Holidays/days that the University is closed by custom (including Christmas) – That’s 42 days a year!
• Generous pension scheme plus life assurance – The University contributes 14.5% of salary.
• Health and Wellbeing: Discounted staff membership options at The Edge, our state-of-the-art Campus gym, with a pool, sauna, climbing wall, cycle circuit, and sports halls.
• Personal Development: Access to courses run by our Organisational Development & Professional Learning team.
• Access to on-site childcare, shopping discounts and travel schemes are also available.

And much more!

If you are looking for a role where you develop real-world impact from your climate dynamics expertise, apply today.

To explore the post further or for any queries you may have, please contact:

Professor Douglas Parker

Email: d.j.parker@leeds.ac.uk

Advert information