Postdoctoral Researcher in Weather and Seasonal prediction
University of Tokyo
Homepage: https://www.t.u-tokyo.ac.jp/en/
Hydrological Sciences (HS)
Nonlinear Processes in Geosciences (NP)
Research and Development of Next-Generation AI-Based Weather Forecast Post-Processing and Downscaling
The successful candidate will conduct research and development on next-generation weather forecast post-processing and downscaling methods. The methods will use forecasts from both physics-based and AI-based global weather prediction models as input and generate low-bias predictions of near-surface atmospheric and marine conditions at spatial scales finer than those of the original global models.
The research will primarily focus on machine-learning-based approaches and aims to establish a post-processing and downscaling framework that satisfies the following three requirements: (1) flexibility to accommodate forecasts from a wide range of global weather prediction models; (2) low computational costs for both model training and inference; and (3) high predictive accuracy for high-impact weather events that significantly affect human activities.
The ultimate goal is to develop a methodology that can become the next-generation de facto standard for weather forecast post-processing and downscaling.
See the activities of our research group from https://sites.google.com/view/sawadaresearchgroup/home
Scope of change: The appointee may be assigned other tasks, hold concurrent positions, or be seconded to other organizations.