Two researchers in Climate Modeling with AI

Lunds universitet, institutionen för naturgeografi och ekosystemvetenskap

Lund University was founded in 1666 and is repeatedly ranked among the world’s top universities. The University has around 45 000 students and more than 8 000 staff based in Lund, Helsingborg and Malmö. We are united in our efforts to understand, explain and improve our world and the human condition.

Lund University welcomes applicants with diverse backgrounds and experiences. We regard gender equality and diversity as a strength and an asset.

Subject description
Climate modelling is within the general field of geobiosphere science and is an aspect of physical geography and ecosystem science.

Clouds are pivotal for the Earth’s radiation budget. This budget determines the extent of climate change for given anthropogenic emissions of greenhouse gases and aerosols. Even an apparently minimal change on average in properties or extent of clouds can cause an appreciable climate change. Climate change itself can alter clouds. The cloud-radiation feedbacks govern the climate sensitivity to radiative forcings, such as greenhouse gas emissions.

Numerical global models of the atmosphere are used to predict climate change. Conventionally, such climate models have treated clouds statistically with 'parameterisations', the assumptions of which can introduce much uncertainty. Such cloud schemes are the main weakness of climate models.

Consequently, a project at Lund University is oriented towards improving the accuracy of climate prediction by use of AI techniques. Machine learning offers a way to simulate the properties and extents of clouds more realistically. The project will advance understanding of the cloud-radiation feedback in climate change. This feedback controls the extent of projected global warming.

Two postdoctoral research positions for one year each are to be filled in the project.

Work duties

Here is a detailed description of the work duties:

• Gather observations of cloud-related quantities and compare these with an AI-based climate model;
• Infer cloud-radiation feedbacks from a present-day simulation with the AI-based climate model;
• Assist with training of the AI-scheme for clouds (e.g. for radiation, global training dataset);
• Include electrical quantities in the AI-scheme for clouds and improve a high-resolution cloud model’s prediction of storm electrification.
• Write a report at the end of the year.

Qualification requirements
Researchers with a background in numerical modeling and knowledge of mesoscale meteorology are encouraged to apply.

Applicants must have:

• PhD degree in meteorology or equivalent
• BSc in physics or equivalent
• Oral and written proficiency in English.
• Programming language experience in UNIX environments (e.g. Linux), with Fortran and python languages.

Assessment criteria and other qualifications

• documented knowledge, preferably from his / her university education in:
  • o mathematics, especially differential equations;
  • o numerical methods and / or computer programming; and
  • o physical meteorology (for example, cloud physics).
• Experience with artificial intelligence software would be an advantage.

Consideration will also be given to good collaborative skills, drive and independence, and how the applicant’s experience and skills complement and strengthen ongoing research within the department, and how they stand to contribute to its future development.

Terms of employment
The employment for each position is for one year. At the end of the year, the results should be summarized in a written report from each researcher.

Both positions will start on 15 December 2023 or at a mutually agreed date and last for one year. The employment will be full-time (100% FTE).

Enquiries about the positions can be made to Vaughan Phillips (vaughan.phillips@nateko.lu.se).

Instructions on how to apply
Applications shall be written in English and be compiled into a PDF-file containing:

• résumé/CV, including a list of publications,
• a general description of past research and future research interests (no more than three pages),
• contact information of at least two references,
• copy of the doctoral degree certificate, and other certificates/grades that you wish to be considered.