408 BOUNDARIES OF THE SEA 



times. It seems unlikely that the atmosphere itself will reveal any 

 processes having response times of the order of a few years, as is 

 required when considering couplings with the ocean surface water, 

 or of several hundred years when considering couplings with the 

 deep sea. One may attain sufficiently large time constants by 

 including the ice in the polar regions. The essential coupling would 

 then be between the ice and the sea, the atmosphere serving as 

 the communicating energy link. It seems possible that the deep 

 sea could be involved in oscillations caused by such a coupling. 

 However, it must be emphasized that one cannot expect such 

 oscillations to give rise to larger temperature fluctuations in the 

 atmosphere, but rather that the fluctuations must be within the 

 same range as those occurring in the deep sea. The reason for this 

 is the enormous damping effect of the deep sea. 



An intensive heating of the sea from the atmosphere, even over 

 a period of a few hundred or a thousand years that corresponds to 

 the circulation time of the deep sea, would produce only small 

 changes in the mean temperature of the sea. The heat stored in 

 the deep sea can be returned to the atmosphere at a later time, 

 but the "quality" of this heat is then much lower, since it would 

 be obtained under conditions with small temperature differences, 

 and cannot be used so effectively. 



For this and also for other reasons it is not very likely that the 

 major climatic changes on the earth that have been recorded can 

 be explained as a self-sustained oscillation. It should also be 

 pointed out that the explanation of these changes based on a 

 variable solar constant has very strong support from the recent 

 detection of an increase in the solar constant by measurement of 

 the light reflected from other planets. 



Now, even though the feedback links between the atmosphere 

 and the ocean do not produce any dramatic instabilities, they 

 may be of importance. To make good long-range weather forecasts 

 the thermal effects from the sea must be included, and the feedback 

 effects will then also have to be considered. For this reason it is uf 

 interest to study further the characteristic responses of the sea to 

 different mechanical and thermodynamic perturbations. 



It would of course also be of the greatest value if one could 



