Postdoctoral Research Associate

Faculty of Environment, Science and Economy

The above full-time post is available on a fixed term basis from the 1^st of April 2025 to 31^st of October 2025 in the Faculty for Environment, Science and Economy.

The post

The Faculty wishes to recruit a Postdoctoral Research Associate to support the work of Stefan Siegert, Frank Kwasniok and Christopher Ferro in collaboration with the UK Met Office. This Met Office funded post is available from 1 April 2025 to 31 October 2025. The successful applicant will explore the integration of AI-based weather prediction models into the Met Office’s post-processing system IMPROVER for blending probabilistic weather forecasts. The project will assess whether and how incorporating AI outputs and machine learning-based blending methods can enhance forecast accuracy and reliability. Specifically, the successful applicant will: Collect and process AI and numerical weather prediction (NWP) model data; Analyse and refine the current blending methodology within IMPROVER; Implement and evaluate AI-enhanced blending techniques; Collaborate with Met Office scientists and external partners; Document findings in reports and contribute to journal publications.

The post will include: Data Collection & Processing: Gathering AI and numerical weather prediction (NWP) model outputs, organising datasets, and preparing data for blending within the IMPROVER system. Model Familiarisation & Analysis: Studying the current IMPROVER blending approach, reviewing relevant literature, and identifying potential improvements. Forecast Blending & Evaluation: Implementing AI-enhanced blending methods, generating probabilistic forecasts, and comparing performance against traditional approaches. Machine Learning (ML) Integration: Exploring advanced ML techniques for blending forecasts and assessing their impact on accuracy. Collaboration & Reporting: Engaging with Met Office scientists and external partners, participating in regular progress meetings, and documenting findings in reports and presentations.

About you

The successful applicant will be able to present information on research progress and outcomes, communicate complex information, orally, in writing and electronically and prepare proposals and applications to external bodies.

Applicants will possess a relevant PhD (or nearing completion) or possess an equivalent qualification/experience in a related field of study and be able to demonstrate sufficient knowledge in the discipline and of research methods and techniques to work within established research programmes. Applicants will have the following essential skills: Analysing high-dimensional data sets from numerical weather prediction (NWP) models and atmospheric observations/reanalyses; Using high-level programming languages and relevant data science libraries for data processing and modeling;   Analysing and evaluating model performance using suitable metrics; Engaging with open-source codebases and adapting existing methodologies for research; Communicating scientific concepts effectively through reports, presentations, and publications; Collaborating with interdisciplinary teams; Proactively managing communications between project partners; Working independently and accurately. The following skills are desirable: Python programming; Experience with AI NWP foundation models; Applying ML methods to meteorological data; ML methodology for post-processing forecasts and blending output from different models.

Please ensure you read the Job Description and Person Specification for full details of this role. 

Further information

For further information please contact Stefan Siegert, e-mail s.siegert@exeter.ac.uk or telephone (01392) 724058.

The closing date for completed applications is the 2^nd of March 2025. Interviews are expected to take place on the 10^th of March 2025.

Advert information