JAMSTEC > Basic Research Area > Department of Seamless Environmental Prediction Research > Global cloud-resolving modeling research

Global cloud-resolving modeling research

Global cloud-resolving modeling research

Complex interactions across physical processes with different spatiotemporal scales cause extreme weather phenomena such as heavy rainfall, tropical cyclone, and intra-seasonal oscillation including Madden-Julian Oscillation (MJO). Therefore, numerical model which can express all elements of "multi-scale phenomena" --- ranging from planetary scale (~10,000 km) to small scale such as individual cloud and turbulence (~10 m) --- is needed to predict/project these weather phenomena. Developing such numerical model requires fundamental understanding of these processes as well as advanced computational technique.
We aim to construct "seamless environmental prediction system", which treats weather forecast, seasonal prediction and climate projection in a united framework, in order to understand and predict weather phenomena over a broad range of spatiotemporal scales. To this end, we are accelerating the fundamental development and sophistication of the numerical model, NICAM (a global cloud-resolving model), as well as ultra-fine mesh process-resolving model and atmosphere-ocean coupled model. In collaboration with ship and satellite observations, we pursue unified understanding of the atmospheric phenomena to improve prediction skill. Development of the visualization technique is also our challenge to express a result effectively.
As a world-leading research group of this field, we collaborate with domestic and international institutes, discuss development and improvement of the model, and introduce advanced technique using the Earth Simulator and the K computer.

Footnote:
1: 30-60 day variability of the large-scale convection activity in the tropics.