Don't miss our weekly PhD newsletter | Sign up now Don't miss our weekly PhD newsletter | Sign up now

  Probabilistic Machine Learning in Climate Science

   School of Electronic Engineering and Computer Science

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Dr Rendani Mbuvha  No more applications being accepted  Funded PhD Project (Students Worldwide)

About the Project

About the Project

Applications are invited for a full PhD scholarship starting Sep 2022/Jan 2023 (or as soon as possible thereafter) to undertake research in Probabilistic Machine Learning in Climate Science.

This PhD studentship is part of a project that aims to provide high fidelity localised probabilistic climate projections critical to many societal activities, including climate change detection and attribution, energy system management, public health and food production. Weather forecasts and long-term climate projections are primarily products of large systems of coupled differential equations that make up numerical weather prediction (NWP) and global circulation models (GCMs). Due to the computationally intensive nature of these models, their outputs are often in coarse spatial and temporal resolutions with a typical GCM spatial resolution of 1(111km) around more expansive equatorial areas. This constraint thus necessitates the need for methods that enhance local resolutions of large scale climate models as most practical applications require spatial resolutions less than 10km. Probabilistic Machine Learning methods present a compelling proposition in the area of enhancement of GCM output as they exploit the non-linearities that may exist while also yielding representations of the uncertainty around parameters and predictions.

The studentship will investigate and develop scalable hybrid physical and statistical models for the enhancement of short-term forecasts from NWPs and longer-term projections from GCMs, as well as how this can improve the robustness of climate change detection and attribution studies. The scope of the project is quite broad. Applicants are encouraged to suggest their own interests and refine the research direction accordingly. 

The PhD will be supervised by Dr Rendani Mbuvha and will be based in the Risk and Information Management Group an interdisciplinary group with a strong publication record and high international impact, which is part of the School of Electronic Engineering and Computer Science (, Queen Mary University of London, UK.


All applicants should have a first-class honours degree or equivalent, or an MSc degree in physics, applied mathematics, computer sciences, Earth sciences, or a related field. Applicants should have a good knowledge of English and the ability to express themselves clearly in both written and spoken form. The successful candidate must be strongly motivated to undertake doctoral studies, as well as must have demonstrated the ability to work independently and perform critical analysis. A record of publishing research in international conferences and/or journals is highly desirable, as well as a strong track record of working in international teams.

The essential selection criteria include:

●    Good coding skills in Python, Matlab and/or R.

●    Ability to work independently or as part of a team.

The desirable selection criteria include:

●    Experience and knowledge of machine learning techniques.

●    Experience in modelling geophysical processes.

All nationalities are eligible to apply for this studentship. We offer a 3-year fully-funded PhD studentship supported by Queen Mary University of London, including student fees and a tax-free stipend. In addition to the studentship, we also welcome applications from self-funded students with relevant backgrounds.

To apply, please follow the online instructions specified by the college website for research degrees: Steps 2 onwards are applicable in this case. Please note that we request a ‘Statement of Research Interests. Your statement (no more than 500 words) should answer two questions:

(i) Why are you interested in the topic described above?

(ii) What relevant experience do you have?

In addition to this, we would also like you to submit a sample of your written work. This might be a chapter of your final year or master's dissertation or a published conference or journal paper.

How to apply

In order to submit your online application, you will need to visit the following webpage: Please scroll down the page and click on “PhD Full-time Computer Science - Semester 2 (January Start)”. The successful PhD candidate will be a member of the Risk and Information Management Group. You should mention this in your application.

Applicants interested in the post, seeking further information or feedback on their suitability, are encouraged to contact Dr Rendani Mbuvha at [Email Address Removed] with the subject “Probabilistic Machine Learning in Climate Science”. All applications must be made via the website mentioned above.

The closing date for applications is 10 July 2022.

Interviews are expected to take place in August 2022.

Starting date: Sep 2022/Jan 2023.

Computer Science (8) Environmental Sciences (13)
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

 About the Project