will lead to much greater predictive capability and earlier warning than are available today. This progress will come from a combination of theoretical modeling, computer simulation, and direct measurement, each drawing on the tools of physics and each conducted by researchers schooled in the methods of physics.
Until the 1980s, atmospheric science had concentrated on the theory and practice of weather forecasting, which involved a time scale of 6 to 10 days. Weather forecasting was based on an understanding of the prevailing instability of large-scale, mid-latitude phenomena resulting from an analysis of the Navier-Stokes fluid dynamic equations. In the oceans, meanwhile, the emphasis was on attempting to understand the physical processes that accounted for mass and heat transport in cases such as the Gulf Stream and the circulations of the ocean basins. At that stage of understanding, it was thought that variability in the oceans and variability in the atmosphere were relatively independent of each other on time scales shorter than decades.
More recently, it has been realized that the ocean and the atmosphere are coupled on much shorter time scales. This realization emanated from the developing understanding of the El Niño phenomenon in the Pacific Ocean. A series of positive and negative feedbacks between the ocean and the atmosphere create this phenomenon, an oscillation on a grand scale, which is responsible for an instability of the climate system in the Pacific region. The understanding of this phenomenon, which rests on the joint fluid dynamics of the ocean and the atmosphere, suggests a predictability in the climate system. Predictability has been demonstrated not only on the weather time scale of 6 to 10 days but also on an interannual time scale of 6 months to 1 or 2 years, the time scale of the El Niño-coupled ocean-atmosphere instability. Since the pioneering work on the El Niño phenomenon, it has been shown that the great monsoon systems of the planet are also coupled ocean-atmosphere phenomena on the same time scale, so that their evolution depends on the same joint dynamics and thermodynamics of the atmosphere and ocean.
An ever-larger fraction of the environmental challenges facing humankind consists of problems requiring better management of human activity to reduce its deleterious impact on natural systems. Problems of this kind arise with increasing frequency because of the larger and more prosperous human population. But they can also be addressed with greater success because of our deeper understanding of the affected systems and an improved capacity