tropics (Manabe and Stouffer, 1988; Fawcett et al., 1997) and causing changes in tropical precipitation. In simulations, imposed North Atlantic cooling causes enhanced wind-driven oceanic upwelling in tropical and extratropical regions, which would bring colder waters to the surface and thus might contribute to additional cooling (Ágústsdóttir et al., 1999). There are many possibilities for regime switches lurking in the collective behavior resulting from coupling the Hadley cell to ocean dynamics.
El Niño is an oscillation of the coupled tropical atmosphere-ocean system. The influence of El Niño extends strongly into the extratropics. How much would El Niño change in a warmer or colder climate? Interest in that question has sharpened because there are indications that the character of El Niño events underwent a shift beginning in the 1970s. Carbon-14 data from corals suggest that changes in upwelling and in the source of subsurface water were involved in the shift (Guilderson and Schrag, 1998). Furthermore, comparisons with prehistoric El Niño records recovered from corals from the Last Glacial Maximum and from the previous major interglacial suggest a systematic relation between global conditions and the temporal character and amplitude of El Niño (Hughen et al., 1999; Tudhope et al., 2001).
Changes in El Niño are important in themselves because El Niño is by far the largest interannual climate signal at present and in the past, but it is possible that such changes might also mediate widespread changes in the global climate regime. Tropical transient motions, including those associated with El Niño, affect the water vapor and cloud distribution and hence the global energy budget (Pierrehumbert and Roca, 1998; Pierrehumbert, 1999). Tropical sea-surface temperature fluctuations, in contrast with those in the middle latitudes, cause variations in deep atmospheric heating, in turn giving rise to waves that powerfully communicate their influence to the rest of the planet. Any change in one part of the tropics tends to drag along the temperature of the entire tropical free troposphere owing to the tight coupling enforced by the Hadley and Walker circulations. The tropics are thus a natural candidate to be a “globalizer” of climate influences. The El Niño cycle might also affect transport of freshwater between the Atlantic and Pacific basins through its influence on low-level wind patterns (Latif et al., 2000; Schmittner et al., 2000). In addition, wind shifts affect the patterns of tropical upwelling and subtropical ocean gyres, possibly leading to changes in heat transport out of the tropical oceans. A treatment of some of the factors governing heat transport within the tropics and subtropics