Project Outline

Greetings from the Director

Dear All

Data Assimilation (DA), which is a mathematical technique used to integrate numerical simulations and observational/experimental data, was originally developed in meteorology and oceanography. Nowadays it is applied in various fields such as space, life and industrial sciences. The Institute of Statistical Mathematics (ISM) has been one of the world leaders in DA through governmental projects such as “The Project for Sustainable Coexistence of Human, Nature, and the Earth” as one of the projects of Research Revolution 2002 held by Ministry of Education, Culture, Sports, Science and Technology in FY2002-2006, “Research of Advanced Data Assimilation and Adaptive Numerical Simulations” as one of the projects of Core Research for Evolutional Science and Technology held by Japan Science and Technology Agency in FY2004-2009, and “Next-Generation Integrated Simulation of Living Matter” for developing software to be executed on the next-generation supercomputer “K computer” under supervision of Advanced Institute for Computational Science, RIKEN since FY2006. Though DA is now accepted as an essential technology in many fields of science, sharing perceptions of present problems and future directions of DA among researchers in the relevant community forms an important step in the further advancement of DA.

In order to realize this, the “Next-Generation Simulation NOE” was established in April, 2010 as a branch of the NOE (Network Of Excellence) establishing project promoted by the ISM. The underlying mission of our NOE is to construct next-generation simulation models capable of predicting future states within the framework of Bayesian online estimation of ultra-high dimesional state variables while continuously receiving large volumes of new observational/experimental data. Related research topics are; (1) fundamental research and development of DA algorithms and their implementation on massively parallel computers, (2) research on rapid generation of high-quality pseudo and physical random numbers, (3) development of visualization software providing valuable information that helps to improve simulation models and that conveys to the society various impacts of research results, (4) cloud computing services that make our statistical calculation programs developed for parallel computing available for anyone, and (5) building a cooperative network with other universities and research institutes associated with DA.

We will do our utmost to live up to your expectations. Please keep your eye on our activities.

Sincerely,

Tomoyuki Higuchi, Prof.
Director of Next-Generation Simulation NOE

Director-General of the Institute of Statistical Mathematics
Director of Research and Development Center for Data Assimilation