Recent progress in computer vision has accelerated the development of statistical downscaling, which uses statistical models to improve the spatiotemporal resolution of impactful climate variables, such as extreme temperatures, wind gusts, and precipitation. Machine learning (ML)-based super-resolution algorithms, which learn from data how to best generate high-resolution images from their low-resolution version, are gaining traction because of their improved accuracy and low computational cost once trained. However, they are rarely designed to perform well on extremes, and their robustness is usually only tested in the present climate, where training data are available. These limitations prevent the widespread adoption of modern ML to better constrain uncertainties in the forecasting of local extremes and in the high-resolution projections of climate change.

Qualifications 

• Master’s degree in a quantitative field closely related to machine learning and/or climate science
• Strong background in scientific programming and data science
• Experience in manipulating scientific datasets, ideally including proficiency in Python.
• A solid foundation in applied mathematics and physics. We appreciate a range of competencies, including but not limited to calculus, differential equations, statistics, mechanics, and thermodynamics
• Strong communication skills in English
• Enthusiasm for both atmospheric/climate science and scientific machine learning
• (Preferred but not required) Experience in high-performance computing 
• (Preferred but not required) Background in numerical weather prediction and/or climate change science
• (Preferred but not required) Teaching experience