172 



[chap. 4 



everywhere around the globe. In some areas, such as the eastern Pacific, there 

 is evidence of a cold-core structure, but areas with warm core predominate, 

 particularly in the upper outflow channels which export the converted con- 

 densation products. 



In analyzing the integrals in (27c) and (28c), we need to know the scale of 

 motion effecting the lateral transports, the vertically integrated sums of which 

 we have just deduced. Fortunately, a sizable transport of sensible heat by 

 geostrophic eddies, such as those associated with mid-latitude warm and cold 

 fronts, has not been found ; horizontal temperature and wind fluctuations in the 

 tropics are poorly correlated. Hence the ageostrophic (driven by and parallel 

 to the pressure head) mass circulation must be the mechanism effecting the 

 heat exchange between the equatorial trough zone and higher latitudes. 

 Because of this fact, we can specify the mass circulation in (27c). If we denote 

 means following the boundary with a bar, (27c) becomes 



Q,^ = Q, + LP + Ra = l\ {Cj,T + Agz)Cn{dplg). (27d) 



Jp=1000m6 



Only one boundary, I, appears in this equation because, as already discussed, we 

 assume that no heat flow takes place across the troughline itself. With the 



X. 4- 



E 



I 5- 

 6- 



7- 



-2-10 12 



C„ (m /sec) 



Fig. 34. Calculated mass circulation (ni/sec) through boundary located 10° latitude from 

 equatorial trough on winter side. Positive sign denotes outflow from trough zone. 

 (After Riehl and Malkus, 1958, Fig. 12.) 



values oicpT + Agz at 10° latitude from the troughline from the climatic study, 

 a vertical profile of c„ is fitted in layers of 100-mb thickness, so that a heat 

 transport divergence Qva of 0.94 units is realized. The constraints of mass 

 continuity, known c„ at the surface (-1.5 m/sec) and at 125 mb (zero), are 

 used to obtain the result in Fig. 34, which corresjDonds well with existing 



