As part of the new Centre for Doctoral Training (CDT) in Mathematics for Our Future Climate, we have two PhD position opportunities at the intersection of data assimilation and machine learning at the University of Reading.

The two projects are:

Large ensembles of machine learning forecasts for advanced nonlinear filters in atmospheric data assimilation
Advised by Eviatar Bach, Sarah Dance, and Amos Lawless

Recently, machine learning (ML) weather forecasting models have shown deterministic forecast skill approaching that of physics-based models, at a small fraction of the computational cost. This provides the opportunity to create very large ensembles of ML forecasts, with the potential to improve data assimilation (DA), the process of optimally combining forecasts and observations to improve the accuracy of weather predictions.

Machine Learning Approaches in Bayesian and Ensemble Data Assimilation
Advised by Eviatar Bach and Jochen Broecker

Data assimilation (DA), the process of combining model predictions with observations, is essential for weather forecasting. Computational limitations render typical DA algorithms suboptimal. This project will use machine learning to infer new DA algorithms that are as close to optimality as possible, leveraging variational inference, in order to improve forecasts.

Both projects would have the opportunities to work with operational forecasting centres, the European Centre for Medium-Range Weather Forecasting (ECMWF) and the Met Office, as well as the CDT partners Imperial College London and the University of Southampton.

The positions would start in October 2024, and are for fully funded 4-year PhD studentship. Interested students should apply as soon as possible through the University of Reading’s application page. Applicants should have a strong mathematics background.

Feel free to contact Eviatar Bach (ebach@caltech.edu) with any questions about the projects or application process.