The interaction between the ocean and atmosphere is critical to our understanding of the earth s climate. The ocean has a much greater heat storage capacity than the atmosphere so that for a given change in temperature the top two and half meters of the ocean will store as much heat as the entire atmosphere. The ocean is a giant thermal reservoir that can moderate and limit climatic excursions on the short-term and promote long-term climate changes produced by the deep ocean circulation that carries the effect of climate forcing far into the future. The ocean is also the major source of water in the atmosphere, which largely determines the radiative properties of the atmosphere, and hence the climate-control of the earth system. All of the oceans are at least partly cloud covered. Clouds modify the temperature of the atmosphere and the ocean by absorbing outgoing longwave radiation from the surface and reflecting incoming shortwave radiation. Changes in the temperature or circulation patterns of the ocean and atmosphere may change the amount and type of cloud cover that, in turn, feeds back to the temperature of the ocean and atmosphere with, as yet, uncertain consequences for climate.
A major goal of current research is to understand the physical basis for the variability of the climate on interannual scales and to forecast this variability. Changes within the ocean-atmosphere system affect global sea level, the distribution of sea surface temperature, sea ice concentrations, the heat carried by ocean currents, and the distribution of heat, moisture and momentum in the atmosphere. Short-term climate effects are associated with changes in the mixed layer of the ocean and the boundary layer of the atmosphere.
The maintenance of the western acific warm pool, a region of the warmest water in the open ocean, has been the focus of considerable recent effort [ Webster and Lukas, 1992]. This area is considered to be fundamental to El Ni¤o-Southern Oscillation (ENSO) phenomena in the ocean and atmosphere. Here the ocean and atmosphere are dramatically coupled with the largest precipitation and latent heat release in the atmosphere [ Webster and Lukas, 1992]. An underlying assumption in the tropics is the adjustment of the atmosphere and ocean are closely connected, resulting in rapid feedback from one part of the system to the other. This warm pool is integral to the development of interannual variations within all of the major ocean basins, however, the formation and maintenance of this warm pool is poorly understood, but presumed to be strongly dependent on large-scale ocean-atmosphere interactions.
The memory of the ocean of atmospheric forcing events contributes to long-term climate changes on time scales of centuries. For example, the deepest water in the ocean originates at only a few places in polar regions where sea ice formation causes the deep mixing of saline cold water. This water receives an imprint of the atmospheric climate and trace gases during its formation at the surface and carries this information into the deep ocean. The circulation beneath the mixed layer determines the long-period contribution of the ocean to the climate by introducing a time delay between the formation and subduction of water in one climate region and its emergence through upwelling and mixing with the surface water decades to centuries later.
Present research is directed primarily toward understanding the interaction between the ocean and the atmosphere in climatically sensitive regions. Global-scale studies must rely on sparse in situ data, satellite remote sensing and large-scale coupled ocean-atmosphere models. In each case they must return to the fundamental assumptions of the surface parameterizations discussed above, but now extend this information to large spatial and temporal averages. A comprehensive review of tropical air-sea interaction in general circulation models can be found in Neelin et al. [1992].