AOSC Seminar
March 16, 2017

Data Assimilation Research at RIKEN: Numerical Weather Prediction and Beyond


Takemasa Miyoshi
Advanced Institute for Computational Science, RIKEN
Abstract:  

Data assimilation plays a key role in numerical weather prediction, combining process-based models with real-world data through dynamical systems theory and statistical mathematics. As computing and sensing technologies advance, data assimilation is becoming more popular in broader science and technology fields. At RIKEN, the Japan’s flagship research institute for all science, we have been pioneering the future possibilities of numerical weather prediction by taking advantage of the powerful K computer and Big Data from advanced observing platforms such as the phased array weather radar and Himawari-8 geostationary satellite, at the scales ranging from global to individual storms at the resolution as high as 100 m. The active research initiated a RIKEN President’s initiative to develop an innovation hub centered at data assimilation, connecting the mathematical science, experimental and observational science, and computational science. The RIKEN International Symposium on Data Assimilation (RISDA 2017), just finished on March 2, 2017, was the highlight of the initiative on the data assimilation innovation hub, with a tremendous success. I will talk about some excitement of RIKEN’s activities on data assimilation research from numerical weather prediction and beyond.