Scientific Colloquium
April 15, 2005


The seminal work of V. Bjerknes (1904) laid the foundation for mathematical modeling of atmospheric dynamics based on the laws of physics.  Subsequent works by Richardson (1922), Rossby (1939), Charney (1947) and Phillips (1956) lead to a successful demonstration of prediction of short-term changes in the large scale atmosphere flow pattern by using dynamical models and heralded the modern era of numerical prediction of weather.

This lecture will describe how the major advances in modeling of weather and climate, and phenomenal increases in computing power, and, space and in-situ observations of atmosphere, land and ocean, have lead to the possibility of extending the techniques of short-term weather prediction (1-10 days) to short-term climate prediction (10-1000 days).

The lecture will summarize the major challenges in making useful short-term climate prediction using dynamical models of the coupled climate system. This lecture will attempt to address the question of whether the current inability to make skillful forecasts of short-term climate variations is due to inaccurate models and insufficient observations or is it an indication of the fundamental limits of the predictability of climate.

Finally, the talk will make some proposals for the future pathways to harvest the realizable predictability of climate for the benefit to society.