138 MALKUS [chap. 4 



and be tested by theories of water-mass origin and age, of production of dee]) 

 water and abyssal circulations. Furthermore, the mere removal of water by 

 evaporation and its addition by precipitation have been shown to create 

 sizeable hydrostatic pressure forces and surface currents, analogous to "curl of 

 wind stress". Now that the distribution of evaporative sinks and precipitation 

 sources is well known, the beautiful computational models of Hough (1897) 

 and Goldsbrough (1933) can meaningfully inquire what role water exchange 

 plays in major current dynamics, particularly in the high precipitation zones 

 of the tropics. Evaporation and rain also alter the temperature and salinity of 

 the surface layers, so that improved E — P figures may also give an approach 

 to the stability and internal dynamics of the thermocline region. Comparisons 

 of its structure produced by the given E — P in a static ocean versus the real 

 one would permit assessment of the effects of a given advective and turbulent 

 regime (Stern, 1960). The oceanographic consequences of these water and 

 energy exchanges are discussed further in other chapters of this book. 



These sections, culminating in Table VI, and Figs. 13 and 15 answer the 

 questions of "what?", "where?" and "how much?" concerning annual global 

 energy transactions, and the interaction between sea and air. We can calculate 

 the amount, origin and form of the energy entering the moving parts of the 

 system — the planetary fluids of sea and air. As a result we have been able to 

 estimate quantitatively in Table I (and test partially. Table IX) the average 

 transports by circulations, and make a first endeavor to break these down 

 between sea and air and into the type of energy transported. 



How these circulations are driven, how the thermal energy is converted into 

 motion on the many scales, and how the motions in turn regulate the energy 

 releases and transports, constitute the basic problem of geophysical fluid 

 dynamics. It is fundamentally a nonlinear problem, or set of problems, in 

 irreversible turbulent fluid thermodynamics, and it is an understatement to 

 say that it has not yet been formally solved on a global scale. Only the simplest 

 "prototype" fluid heat engines are as yet tractable to rigorous mathematical 

 analysis (cf. W. V. R. Malkus, 1954). On the global scale, exciting beginnings of 

 general circulation studies have been undertaken by numerical and experi- 

 mental methods for both atmosphere (Lorenz, 1960; Phiflips, 1956; Fultz, 

 1956; Riehl and Fultz, 1957, 1958) and ocean (Stommel, Arons and Faller, 

 1958). As we shall see in Sections 5 and 6, progress in semi-theoretical modelling 

 of the conversion of thermal energy into motion has been made for some of the 

 component regions and scales in the atmosphere, particularly in the tropics. 



To progress in this direction, the mean annual picture of energy exchanges 

 and transportation provide a beginning framework, particularly as a foundation 

 and proving ground for steady-state circulation theories. We may inquire what 

 average or integrated motion scheme is consistent with the energy sources, 

 sinks, exports, imports and internal releases that have been deduced quantita- 

 tively herein. In order to approach the vitally important fluctuations and 

 instabilities in the air-sea system, however, we must examine the time depend- 

 ences of these energy transactions, and how the various interacting scales of 



