Chiba University
Just Landed
Posted Yesterday
Specially Appointed Researcher/Postdoctoral Position in AI-Based Intraseasonal and Seasonal Weather Prediction at Chiba University Chiba University in Japan
Degree Level
Postdoc
Field of study
Computer Science
Funding
Annual salary is approximately 5,000,000–6,000,000 Japanese yen, determined according to Chiba University's regulations and experience; approximately 5,500,000 yen is expected for a researcher in the first year after obtaining the degree. The position includes commuting allowance subject to university regulations and enrollment in health insurance, employees' pension, workers' compensation insurance, and employment insurance. This is a fixed-term, annually renewable contract position without ten
Deadline
Dec 25, 2026
Country
Japan
University
Chiba University
Apply for this position
Keywords
Suggested positions
About this position
Chiba University's Center for Environmental Remote Sensing (CEReS) is seeking a specially appointed researcher to develop AI-based intraseasonal and seasonal prediction models. The project targets forecasts from approximately 10 to 90 days and aims to improve their usefulness for social applications such as ship operations.
The researcher will work on AI weather and climate modeling that combines machine learning with ensemble data assimilation and physically informed forecasting. The project uses systems including ClimaX-LETKF, which combines an AI weather forecasting model with ensemble data assimilation, and Anemoi-based global and regional graph neural networks. A central goal is to develop models that remain numerically stable and physically consistent during integrations extending up to three months.
Potential research directions include adapting existing AI forecasting systems for intraseasonal and seasonal prediction; evaluating architectures such as Aurora, SFNO, and Anemoi; developing models trained on daily or longer-timescale averages; correcting forecast drift and error with conditional diffusion models, generative AI, or statistical methods; and applying Transformer-based super-resolution to low-resolution physical-model and AI-model outputs.
The broader research will integrate AI forecasting, generative models, data assimilation, and ensemble prediction while representing atmospheric and oceanic processes relevant to seasonal predictability and remote influences. The researcher will publish academic papers, present results at domestic and international conferences, and collaborate with participating institutions.
The work is part of a Japan Science and Technology Agency project on a meteorological and oceanographic seasonal prediction simulator, led by Yasumasa Miyazawa of JAMSTEC, and will be conducted in collaboration with the Mathematical Flood Science Research Group at the University of Tokyo, led by PI Yohei Sawada. The successful researcher will join CEReS at Chiba University's campus in Chiba, Japan.
Funding details
Annual salary is approximately 5,000,000–6,000,000 Japanese yen, determined according to Chiba University's regulations and experience; approximately 5,500,000 yen is expected for a researcher in the first year after obtaining the degree. The position includes commuting allowance subject to university regulations and enrollment in health insurance, employees' pension, workers' compensation insurance, and employment insurance. This is a fixed-term, annually renewable contract position without ten
What's required
Applicants must hold a doctoral degree or be expected to obtain one in the near future. Applicants should have research experience, or a strong interest, in one or more of machine learning, deep learning, generative AI, meteorology, climate science, atmospheric science, oceanography, numerical weather prediction, seasonal prediction, data assimilation, remote sensing, applied mathematics, information science, or data science. Experience developing AI models or conducting large-scale numerical computations in information science, mathematical science, computational science, physics, or engineering is welcome, including for applicants seeking to transition into weather and climate prediction. Applicants should be able to contribute to research on stable long-term AI forecasting, integration of AI models with data assimilation, forecast correction using conditional diffusion models, and super-resolution of low-resolution prediction data. The application requires a CV, publication list, three major-paper reprints, a research history and proposed research plan including usable programming languages, and contact details for one or two referees outside the center.
How to apply
Download the designated Chiba University CV form from the CEReS website. Mail the CV, publication list, three major-paper reprints, research history and post-appointment research plan, and one or two external referee contacts by registered mail to Shunji Kotsuki at the Center for Environmental Remote Sensing. Write the specified application phrase on the envelope; electronic and JREC-IN applications are not accepted.
Make your next move.
Prepare your application
Review drafts and confirm requirements with the university.
Make your next move.
Prepare your application
Review drafts and confirm requirements with the university.